diff --git a/llama2_13b_peft/paragraph_segmentation/README.md b/llama2_13b_peft/paragraph_segmentation/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..5d63cc331a367c0986e1e9aa53333bd2232856b2
--- /dev/null
+++ b/llama2_13b_peft/paragraph_segmentation/README.md
@@ -0,0 +1,94 @@
+---
+license: other
+library_name: peft
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+base_model: /data1/model/llama2/meta-llama/Llama2-13b
+model-index:
+- name: paragraph_segmentation_no_sys
+ results: []
+---
+
+
+
+# paragraph_segmentation_no_sys
+
+This model is a fine-tuned version of [/data1/model/llama2/meta-llama/Llama2-13b](https://huggingface.co//data1/model/llama2/meta-llama/Llama2-13b) on the paragraph_segmentation_no_sys dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.1180
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 4
+- eval_batch_size: 4
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 2
+- total_train_batch_size: 8
+- total_eval_batch_size: 8
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_steps: 20
+- num_epochs: 5.0
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:------:|:----:|:---------------:|
+| 0.2344 | 0.1170 | 100 | 0.2354 |
+| 0.2247 | 0.2339 | 200 | 0.2296 |
+| 0.2009 | 0.3509 | 300 | 0.2020 |
+| 0.2097 | 0.4678 | 400 | 0.1956 |
+| 0.185 | 0.5848 | 500 | 0.1874 |
+| 0.1804 | 0.7018 | 600 | 0.1784 |
+| 0.1647 | 0.8187 | 700 | 0.1701 |
+| 0.1663 | 0.9357 | 800 | 0.1684 |
+| 0.1553 | 1.0526 | 900 | 0.1524 |
+| 0.1665 | 1.1696 | 1000 | 0.1584 |
+| 0.1424 | 1.2865 | 1100 | 0.1660 |
+| 0.1163 | 1.4035 | 1200 | 0.1442 |
+| 0.1593 | 1.5205 | 1300 | 0.1365 |
+| 0.1529 | 1.6374 | 1400 | 0.1444 |
+| 0.1321 | 1.7544 | 1500 | 0.1334 |
+| 0.135 | 1.8713 | 1600 | 0.1293 |
+| 0.1322 | 1.9883 | 1700 | 0.1401 |
+| 0.1355 | 2.1053 | 1800 | 0.1292 |
+| 0.1227 | 2.2222 | 1900 | 0.1354 |
+| 0.1061 | 2.3392 | 2000 | 0.1234 |
+| 0.1164 | 2.4561 | 2100 | 0.1263 |
+| 0.1038 | 2.5731 | 2200 | 0.1236 |
+| 0.1196 | 2.6901 | 2300 | 0.1217 |
+| 0.1058 | 2.8070 | 2400 | 0.1206 |
+| 0.1177 | 2.9240 | 2500 | 0.1204 |
+| 0.1009 | 3.0409 | 2600 | 0.1180 |
+| 0.1024 | 3.1579 | 2700 | 0.1185 |
+| 0.0953 | 3.2749 | 2800 | 0.1236 |
+| 0.0905 | 3.3918 | 2900 | 0.1202 |
+
+
+### Framework versions
+
+- PEFT 0.10.0
+- Transformers 4.40.0
+- Pytorch 2.2.1
+- Datasets 2.18.0
+- Tokenizers 0.19.1
\ No newline at end of file
diff --git a/llama2_13b_peft/paragraph_segmentation/adapter_config.json b/llama2_13b_peft/paragraph_segmentation/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..b36ad376f9ea0b5fdb04d70e360d6df8e9980f80
--- /dev/null
+++ b/llama2_13b_peft/paragraph_segmentation/adapter_config.json
@@ -0,0 +1,34 @@
+{
+ "alpha_pattern": {},
+ "auto_mapping": null,
+ "base_model_name_or_path": "/data1/model/llama2/meta-llama/Llama2-13b",
+ "bias": "none",
+ "fan_in_fan_out": false,
+ "inference_mode": true,
+ "init_lora_weights": true,
+ "layer_replication": null,
+ "layers_pattern": null,
+ "layers_to_transform": null,
+ "loftq_config": {},
+ "lora_alpha": 16,
+ "lora_dropout": 0.0,
+ "megatron_config": null,
+ "megatron_core": "megatron.core",
+ "modules_to_save": null,
+ "peft_type": "LORA",
+ "r": 8,
+ "rank_pattern": {},
+ "revision": null,
+ "target_modules": [
+ "k_proj",
+ "gate_proj",
+ "down_proj",
+ "o_proj",
+ "up_proj",
+ "v_proj",
+ "q_proj"
+ ],
+ "task_type": "CAUSAL_LM",
+ "use_dora": false,
+ "use_rslora": false
+}
\ No newline at end of file
diff --git a/llama2_13b_peft/paragraph_segmentation/adapter_model.safetensors b/llama2_13b_peft/paragraph_segmentation/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..e111b0e8e0cc13a010d4438d6e1a2180cae08580
--- /dev/null
+++ b/llama2_13b_peft/paragraph_segmentation/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f4aac6fb5909f3aa5cf0aeb90a4e4e745847e28d213cda4ccbdd852fe680ae7b
+size 125248064
diff --git a/llama2_13b_peft/paragraph_segmentation/all_results.json b/llama2_13b_peft/paragraph_segmentation/all_results.json
new file mode 100644
index 0000000000000000000000000000000000000000..bf69ae9ea067c1102a92934aa40897244cb510b8
--- /dev/null
+++ b/llama2_13b_peft/paragraph_segmentation/all_results.json
@@ -0,0 +1,12 @@
+{
+ "epoch": 3.391812865497076,
+ "eval_loss": 0.11801114678382874,
+ "eval_runtime": 70.3409,
+ "eval_samples_per_second": 5.118,
+ "eval_steps_per_second": 0.64,
+ "total_flos": 5.067618607815459e+18,
+ "train_loss": 0.14973004334959492,
+ "train_runtime": 17301.2227,
+ "train_samples_per_second": 1.977,
+ "train_steps_per_second": 0.247
+}
\ No newline at end of file
diff --git a/llama2_13b_peft/paragraph_segmentation/eval_results.json b/llama2_13b_peft/paragraph_segmentation/eval_results.json
new file mode 100644
index 0000000000000000000000000000000000000000..08a07585d2bda993ccfd75bced4acd14c1cc8f1b
--- /dev/null
+++ b/llama2_13b_peft/paragraph_segmentation/eval_results.json
@@ -0,0 +1,7 @@
+{
+ "epoch": 3.391812865497076,
+ "eval_loss": 0.11801114678382874,
+ "eval_runtime": 70.3409,
+ "eval_samples_per_second": 5.118,
+ "eval_steps_per_second": 0.64
+}
\ No newline at end of file
diff --git a/llama2_13b_peft/paragraph_segmentation/special_tokens_map.json b/llama2_13b_peft/paragraph_segmentation/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..72ecfeeb7e14d244c936169d2ed139eeae235ef1
--- /dev/null
+++ b/llama2_13b_peft/paragraph_segmentation/special_tokens_map.json
@@ -0,0 +1,24 @@
+{
+ "bos_token": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false
+ },
+ "eos_token": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false
+ },
+ "pad_token": "",
+ "unk_token": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false
+ }
+}
diff --git a/llama2_13b_peft/paragraph_segmentation/tokenizer.model b/llama2_13b_peft/paragraph_segmentation/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..6c00c742ce03c627d6cd5b795984876fa49fa899
--- /dev/null
+++ b/llama2_13b_peft/paragraph_segmentation/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723
diff --git a/llama2_13b_peft/paragraph_segmentation/tokenizer_config.json b/llama2_13b_peft/paragraph_segmentation/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..a40266f39e5b5fed14de34710d35eb9e98d6bdad
--- /dev/null
+++ b/llama2_13b_peft/paragraph_segmentation/tokenizer_config.json
@@ -0,0 +1,45 @@
+{
+ "add_bos_token": true,
+ "add_eos_token": false,
+ "add_prefix_space": true,
+ "added_tokens_decoder": {
+ "0": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false,
+ "special": true
+ },
+ "1": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false,
+ "special": true
+ },
+ "2": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false,
+ "special": true
+ }
+ },
+ "bos_token": "",
+ "chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '\\n' }}{% endif %}{% endfor %}",
+ "clean_up_tokenization_spaces": false,
+ "eos_token": "",
+ "legacy": true,
+ "model_max_length": 1000000000000000019884624838656,
+ "pad_token": "",
+ "padding_side": "right",
+ "sp_model_kwargs": {},
+ "spaces_between_special_tokens": false,
+ "split_special_tokens": false,
+ "tokenizer_class": "LlamaTokenizer",
+ "unk_token": "",
+ "use_default_system_prompt": false
+}
diff --git a/llama2_13b_peft/paragraph_segmentation/train_results.json b/llama2_13b_peft/paragraph_segmentation/train_results.json
new file mode 100644
index 0000000000000000000000000000000000000000..0e297662c1167a7e48f90f11a632e3107602658f
--- /dev/null
+++ b/llama2_13b_peft/paragraph_segmentation/train_results.json
@@ -0,0 +1,8 @@
+{
+ "epoch": 3.391812865497076,
+ "total_flos": 5.067618607815459e+18,
+ "train_loss": 0.14973004334959492,
+ "train_runtime": 17301.2227,
+ "train_samples_per_second": 1.977,
+ "train_steps_per_second": 0.247
+}
\ No newline at end of file
diff --git a/llama2_13b_peft/paragraph_segmentation/trainer_log.jsonl b/llama2_13b_peft/paragraph_segmentation/trainer_log.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..cb5a8f3f2b978c412966fc04d4cfe01939292eff
--- /dev/null
+++ b/llama2_13b_peft/paragraph_segmentation/trainer_log.jsonl
@@ -0,0 +1,321 @@
+{"current_steps": 10, "total_steps": 4275, "loss": 0.781, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.5e-05, "epoch": 0.011695906432748537, "percentage": 0.23, "elapsed_time": "0:00:53", "remaining_time": "6:16:58"}
+{"current_steps": 20, "total_steps": 4275, "loss": 0.5723, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 5e-05, "epoch": 0.023391812865497075, "percentage": 0.47, "elapsed_time": "0:01:43", "remaining_time": "6:06:17"}
+{"current_steps": 30, "total_steps": 4275, "loss": 0.378, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.999931858975852e-05, "epoch": 0.03508771929824561, "percentage": 0.7, "elapsed_time": "0:02:38", "remaining_time": "6:13:33"}
+{"current_steps": 40, "total_steps": 4275, "loss": 0.2932, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.999727439617963e-05, "epoch": 0.04678362573099415, "percentage": 0.94, "elapsed_time": "0:03:29", "remaining_time": "6:09:17"}
+{"current_steps": 50, "total_steps": 4275, "loss": 0.2703, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.999386753069811e-05, "epoch": 0.05847953216374269, "percentage": 1.17, "elapsed_time": "0:04:25", "remaining_time": "6:14:10"}
+{"current_steps": 60, "total_steps": 4275, "loss": 0.2844, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.9989098179031804e-05, "epoch": 0.07017543859649122, "percentage": 1.4, "elapsed_time": "0:05:25", "remaining_time": "6:20:48"}
+{"current_steps": 70, "total_steps": 4275, "loss": 0.249, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.998296660117151e-05, "epoch": 0.08187134502923976, "percentage": 1.64, "elapsed_time": "0:06:09", "remaining_time": "6:10:23"}
+{"current_steps": 80, "total_steps": 4275, "loss": 0.2521, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.997547313136682e-05, "epoch": 0.0935672514619883, "percentage": 1.87, "elapsed_time": "0:07:05", "remaining_time": "6:12:14"}
+{"current_steps": 90, "total_steps": 4275, "loss": 0.2647, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.996661817810791e-05, "epoch": 0.10526315789473684, "percentage": 2.11, "elapsed_time": "0:07:59", "remaining_time": "6:11:55"}
+{"current_steps": 100, "total_steps": 4275, "loss": 0.2344, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.995640222410324e-05, "epoch": 0.11695906432748537, "percentage": 2.34, "elapsed_time": "0:08:47", "remaining_time": "6:07:16"}
+{"current_steps": 100, "total_steps": 4275, "loss": null, "eval_loss": 0.23539581894874573, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 0.11695906432748537, "percentage": 2.34, "elapsed_time": "0:08:47", "remaining_time": "6:07:16"}
+{"current_steps": 110, "total_steps": 4275, "loss": 0.2274, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.994482582625327e-05, "epoch": 0.1286549707602339, "percentage": 2.57, "elapsed_time": "0:10:46", "remaining_time": "6:47:48"}
+{"current_steps": 120, "total_steps": 4275, "loss": 0.2664, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.9931889615620075e-05, "epoch": 0.14035087719298245, "percentage": 2.81, "elapsed_time": "0:11:42", "remaining_time": "6:45:36"}
+{"current_steps": 130, "total_steps": 4275, "loss": 0.2483, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.991759429739298e-05, "epoch": 0.15204678362573099, "percentage": 3.04, "elapsed_time": "0:12:36", "remaining_time": "6:41:56"}
+{"current_steps": 140, "total_steps": 4275, "loss": 0.2269, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.9901940650850074e-05, "epoch": 0.16374269005847952, "percentage": 3.27, "elapsed_time": "0:13:27", "remaining_time": "6:37:32"}
+{"current_steps": 150, "total_steps": 4275, "loss": 0.2249, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.9884929529315774e-05, "epoch": 0.17543859649122806, "percentage": 3.51, "elapsed_time": "0:14:21", "remaining_time": "6:34:55"}
+{"current_steps": 160, "total_steps": 4275, "loss": 0.2232, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.986656186011426e-05, "epoch": 0.1871345029239766, "percentage": 3.74, "elapsed_time": "0:15:21", "remaining_time": "6:35:11"}
+{"current_steps": 170, "total_steps": 4275, "loss": 0.2274, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.984683864451898e-05, "epoch": 0.19883040935672514, "percentage": 3.98, "elapsed_time": "0:16:10", "remaining_time": "6:30:41"}
+{"current_steps": 180, "total_steps": 4275, "loss": 0.2153, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.982576095769801e-05, "epoch": 0.21052631578947367, "percentage": 4.21, "elapsed_time": "0:17:06", "remaining_time": "6:29:21"}
+{"current_steps": 190, "total_steps": 4275, "loss": 0.2123, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.980332994865549e-05, "epoch": 0.2222222222222222, "percentage": 4.44, "elapsed_time": "0:17:56", "remaining_time": "6:25:36"}
+{"current_steps": 200, "total_steps": 4275, "loss": 0.2247, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.9779546840168964e-05, "epoch": 0.23391812865497075, "percentage": 4.68, "elapsed_time": "0:18:50", "remaining_time": "6:23:50"}
+{"current_steps": 200, "total_steps": 4275, "loss": null, "eval_loss": 0.2295626550912857, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 0.23391812865497075, "percentage": 4.68, "elapsed_time": "0:18:50", "remaining_time": "6:23:50"}
+{"current_steps": 210, "total_steps": 4275, "loss": 0.2358, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.9754412928722714e-05, "epoch": 0.24561403508771928, "percentage": 4.91, "elapsed_time": "0:20:47", "remaining_time": "6:42:19"}
+{"current_steps": 220, "total_steps": 4275, "loss": 0.2187, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.9727929584437136e-05, "epoch": 0.2573099415204678, "percentage": 5.15, "elapsed_time": "0:21:46", "remaining_time": "6:41:18"}
+{"current_steps": 230, "total_steps": 4275, "loss": 0.2077, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.970009825099398e-05, "epoch": 0.26900584795321636, "percentage": 5.38, "elapsed_time": "0:22:37", "remaining_time": "6:37:51"}
+{"current_steps": 240, "total_steps": 4275, "loss": 0.2458, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.967092044555769e-05, "epoch": 0.2807017543859649, "percentage": 5.61, "elapsed_time": "0:23:25", "remaining_time": "6:33:46"}
+{"current_steps": 250, "total_steps": 4275, "loss": 0.1873, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.9640397758692715e-05, "epoch": 0.29239766081871343, "percentage": 5.85, "elapsed_time": "0:24:22", "remaining_time": "6:32:19"}
+{"current_steps": 260, "total_steps": 4275, "loss": 0.1758, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.960853185427676e-05, "epoch": 0.30409356725146197, "percentage": 6.08, "elapsed_time": "0:25:18", "remaining_time": "6:30:42"}
+{"current_steps": 270, "total_steps": 4275, "loss": 0.2229, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.957532446941012e-05, "epoch": 0.3157894736842105, "percentage": 6.32, "elapsed_time": "0:26:14", "remaining_time": "6:29:08"}
+{"current_steps": 280, "total_steps": 4275, "loss": 0.2043, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.954077741432097e-05, "epoch": 0.32748538011695905, "percentage": 6.55, "elapsed_time": "0:27:10", "remaining_time": "6:27:47"}
+{"current_steps": 290, "total_steps": 4275, "loss": 0.1949, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.950489257226667e-05, "epoch": 0.3391812865497076, "percentage": 6.78, "elapsed_time": "0:28:01", "remaining_time": "6:25:09"}
+{"current_steps": 300, "total_steps": 4275, "loss": 0.2009, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.9467671899431145e-05, "epoch": 0.3508771929824561, "percentage": 7.02, "elapsed_time": "0:28:53", "remaining_time": "6:22:51"}
+{"current_steps": 300, "total_steps": 4275, "loss": null, "eval_loss": 0.20195595920085907, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 0.3508771929824561, "percentage": 7.02, "elapsed_time": "0:28:53", "remaining_time": "6:22:51"}
+{"current_steps": 310, "total_steps": 4275, "loss": 0.1841, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.942911742481819e-05, "epoch": 0.36257309941520466, "percentage": 7.25, "elapsed_time": "0:30:56", "remaining_time": "6:35:48"}
+{"current_steps": 320, "total_steps": 4275, "loss": 0.2091, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.938923125014094e-05, "epoch": 0.3742690058479532, "percentage": 7.49, "elapsed_time": "0:31:49", "remaining_time": "6:33:14"}
+{"current_steps": 330, "total_steps": 4275, "loss": 0.2032, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.93480155497072e-05, "epoch": 0.38596491228070173, "percentage": 7.72, "elapsed_time": "0:32:44", "remaining_time": "6:31:27"}
+{"current_steps": 340, "total_steps": 4275, "loss": 0.1952, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.930547257030102e-05, "epoch": 0.39766081871345027, "percentage": 7.95, "elapsed_time": "0:33:39", "remaining_time": "6:29:37"}
+{"current_steps": 350, "total_steps": 4275, "loss": 0.1926, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.926160463106015e-05, "epoch": 0.4093567251461988, "percentage": 8.19, "elapsed_time": "0:34:37", "remaining_time": "6:28:14"}
+{"current_steps": 360, "total_steps": 4275, "loss": 0.2085, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.921641412334963e-05, "epoch": 0.42105263157894735, "percentage": 8.42, "elapsed_time": "0:35:32", "remaining_time": "6:26:26"}
+{"current_steps": 370, "total_steps": 4275, "loss": 0.199, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.9169903510631434e-05, "epoch": 0.4327485380116959, "percentage": 8.65, "elapsed_time": "0:36:31", "remaining_time": "6:25:25"}
+{"current_steps": 380, "total_steps": 4275, "loss": 0.1874, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.91220753283302e-05, "epoch": 0.4444444444444444, "percentage": 8.89, "elapsed_time": "0:37:28", "remaining_time": "6:24:06"}
+{"current_steps": 390, "total_steps": 4275, "loss": 0.1775, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.907293218369499e-05, "epoch": 0.45614035087719296, "percentage": 9.12, "elapsed_time": "0:38:22", "remaining_time": "6:22:11"}
+{"current_steps": 400, "total_steps": 4275, "loss": 0.2097, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.9022476755657155e-05, "epoch": 0.4678362573099415, "percentage": 9.36, "elapsed_time": "0:39:19", "remaining_time": "6:20:53"}
+{"current_steps": 400, "total_steps": 4275, "loss": null, "eval_loss": 0.19560903310775757, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 0.4678362573099415, "percentage": 9.36, "elapsed_time": "0:39:19", "remaining_time": "6:20:53"}
+{"current_steps": 410, "total_steps": 4275, "loss": 0.1872, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.8970711794684345e-05, "epoch": 0.47953216374269003, "percentage": 9.59, "elapsed_time": "0:41:17", "remaining_time": "6:29:12"}
+{"current_steps": 420, "total_steps": 4275, "loss": 0.1936, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.8917640122630506e-05, "epoch": 0.49122807017543857, "percentage": 9.82, "elapsed_time": "0:42:09", "remaining_time": "6:26:55"}
+{"current_steps": 430, "total_steps": 4275, "loss": 0.1933, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.886326463258212e-05, "epoch": 0.5029239766081871, "percentage": 10.06, "elapsed_time": "0:43:01", "remaining_time": "6:24:39"}
+{"current_steps": 440, "total_steps": 4275, "loss": 0.1874, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.880758828870045e-05, "epoch": 0.5146198830409356, "percentage": 10.29, "elapsed_time": "0:43:55", "remaining_time": "6:22:51"}
+{"current_steps": 450, "total_steps": 4275, "loss": 0.184, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.875061412605997e-05, "epoch": 0.5263157894736842, "percentage": 10.53, "elapsed_time": "0:44:50", "remaining_time": "6:21:11"}
+{"current_steps": 460, "total_steps": 4275, "loss": 0.1901, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.8692345250482914e-05, "epoch": 0.5380116959064327, "percentage": 10.76, "elapsed_time": "0:45:46", "remaining_time": "6:19:39"}
+{"current_steps": 470, "total_steps": 4275, "loss": 0.1803, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.863278483836996e-05, "epoch": 0.5497076023391813, "percentage": 10.99, "elapsed_time": "0:46:43", "remaining_time": "6:18:18"}
+{"current_steps": 480, "total_steps": 4275, "loss": 0.1688, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.857193613652711e-05, "epoch": 0.5614035087719298, "percentage": 11.23, "elapsed_time": "0:47:39", "remaining_time": "6:16:50"}
+{"current_steps": 490, "total_steps": 4275, "loss": 0.1757, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.8509802461988635e-05, "epoch": 0.5730994152046783, "percentage": 11.46, "elapsed_time": "0:48:38", "remaining_time": "6:15:41"}
+{"current_steps": 500, "total_steps": 4275, "loss": 0.185, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.844638720183632e-05, "epoch": 0.5847953216374269, "percentage": 11.7, "elapsed_time": "0:49:26", "remaining_time": "6:13:20"}
+{"current_steps": 500, "total_steps": 4275, "loss": null, "eval_loss": 0.1873869150876999, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 0.5847953216374269, "percentage": 11.7, "elapsed_time": "0:49:26", "remaining_time": "6:13:20"}
+{"current_steps": 510, "total_steps": 4275, "loss": 0.1864, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.8381693813014775e-05, "epoch": 0.5964912280701754, "percentage": 11.93, "elapsed_time": "0:51:29", "remaining_time": "6:20:08"}
+{"current_steps": 520, "total_steps": 4275, "loss": 0.1723, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.8315725822143024e-05, "epoch": 0.6081871345029239, "percentage": 12.16, "elapsed_time": "0:52:23", "remaining_time": "6:18:19"}
+{"current_steps": 530, "total_steps": 4275, "loss": 0.161, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.824848682532224e-05, "epoch": 0.6198830409356725, "percentage": 12.4, "elapsed_time": "0:53:16", "remaining_time": "6:16:23"}
+{"current_steps": 540, "total_steps": 4275, "loss": 0.1908, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.817998048793969e-05, "epoch": 0.631578947368421, "percentage": 12.63, "elapsed_time": "0:54:01", "remaining_time": "6:13:43"}
+{"current_steps": 550, "total_steps": 4275, "loss": 0.1888, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.811021054446897e-05, "epoch": 0.6432748538011696, "percentage": 12.87, "elapsed_time": "0:54:45", "remaining_time": "6:10:54"}
+{"current_steps": 560, "total_steps": 4275, "loss": 0.1944, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.803918079826641e-05, "epoch": 0.6549707602339181, "percentage": 13.1, "elapsed_time": "0:55:34", "remaining_time": "6:08:42"}
+{"current_steps": 570, "total_steps": 4275, "loss": 0.1819, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.7966895121363744e-05, "epoch": 0.6666666666666666, "percentage": 13.33, "elapsed_time": "0:56:27", "remaining_time": "6:07:00"}
+{"current_steps": 580, "total_steps": 4275, "loss": 0.177, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.789335745425698e-05, "epoch": 0.6783625730994152, "percentage": 13.57, "elapsed_time": "0:57:19", "remaining_time": "6:05:09"}
+{"current_steps": 590, "total_steps": 4275, "loss": 0.1756, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.781857180569171e-05, "epoch": 0.6900584795321637, "percentage": 13.8, "elapsed_time": "0:58:08", "remaining_time": "6:03:10"}
+{"current_steps": 600, "total_steps": 4275, "loss": 0.1804, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.774254225244447e-05, "epoch": 0.7017543859649122, "percentage": 14.04, "elapsed_time": "0:59:05", "remaining_time": "6:01:57"}
+{"current_steps": 600, "total_steps": 4275, "loss": null, "eval_loss": 0.17837749421596527, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 0.7017543859649122, "percentage": 14.04, "elapsed_time": "0:59:05", "remaining_time": "6:01:57"}
+{"current_steps": 610, "total_steps": 4275, "loss": 0.182, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.766527293910056e-05, "epoch": 0.7134502923976608, "percentage": 14.27, "elapsed_time": "1:01:15", "remaining_time": "6:08:03"}
+{"current_steps": 620, "total_steps": 4275, "loss": 0.1955, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.75867680778281e-05, "epoch": 0.7251461988304093, "percentage": 14.5, "elapsed_time": "1:02:07", "remaining_time": "6:06:16"}
+{"current_steps": 630, "total_steps": 4275, "loss": 0.1685, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.750703194814841e-05, "epoch": 0.7368421052631579, "percentage": 14.74, "elapsed_time": "1:02:52", "remaining_time": "6:03:43"}
+{"current_steps": 640, "total_steps": 4275, "loss": 0.1852, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.7426068896702704e-05, "epoch": 0.7485380116959064, "percentage": 14.97, "elapsed_time": "1:03:38", "remaining_time": "6:01:26"}
+{"current_steps": 650, "total_steps": 4275, "loss": 0.1782, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.734388333701519e-05, "epoch": 0.7602339181286549, "percentage": 15.2, "elapsed_time": "1:04:23", "remaining_time": "5:59:06"}
+{"current_steps": 660, "total_steps": 4275, "loss": 0.1661, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.726047974925244e-05, "epoch": 0.7719298245614035, "percentage": 15.44, "elapsed_time": "1:05:11", "remaining_time": "5:57:04"}
+{"current_steps": 670, "total_steps": 4275, "loss": 0.1638, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.717586267997916e-05, "epoch": 0.783625730994152, "percentage": 15.67, "elapsed_time": "1:06:05", "remaining_time": "5:55:35"}
+{"current_steps": 680, "total_steps": 4275, "loss": 0.1733, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.709003674191035e-05, "epoch": 0.7953216374269005, "percentage": 15.91, "elapsed_time": "1:06:57", "remaining_time": "5:54:02"}
+{"current_steps": 690, "total_steps": 4275, "loss": 0.1798, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.7003006613659866e-05, "epoch": 0.8070175438596491, "percentage": 16.14, "elapsed_time": "1:07:52", "remaining_time": "5:52:39"}
+{"current_steps": 700, "total_steps": 4275, "loss": 0.1647, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.691477703948537e-05, "epoch": 0.8187134502923976, "percentage": 16.37, "elapsed_time": "1:08:46", "remaining_time": "5:51:13"}
+{"current_steps": 700, "total_steps": 4275, "loss": null, "eval_loss": 0.1701304167509079, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 0.8187134502923976, "percentage": 16.37, "elapsed_time": "1:08:46", "remaining_time": "5:51:13"}
+{"current_steps": 710, "total_steps": 4275, "loss": 0.1569, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.6825352829029705e-05, "epoch": 0.8304093567251462, "percentage": 16.61, "elapsed_time": "1:10:49", "remaining_time": "5:55:37"}
+{"current_steps": 720, "total_steps": 4275, "loss": 0.1748, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.673473885705869e-05, "epoch": 0.8421052631578947, "percentage": 16.84, "elapsed_time": "1:11:45", "remaining_time": "5:54:19"}
+{"current_steps": 730, "total_steps": 4275, "loss": 0.2056, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.66429400631954e-05, "epoch": 0.8538011695906432, "percentage": 17.08, "elapsed_time": "1:12:39", "remaining_time": "5:52:50"}
+{"current_steps": 740, "total_steps": 4275, "loss": 0.1772, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.654996145165092e-05, "epoch": 0.8654970760233918, "percentage": 17.31, "elapsed_time": "1:13:33", "remaining_time": "5:51:21"}
+{"current_steps": 750, "total_steps": 4275, "loss": 0.1655, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.645580809095147e-05, "epoch": 0.8771929824561403, "percentage": 17.54, "elapsed_time": "1:14:26", "remaining_time": "5:49:51"}
+{"current_steps": 760, "total_steps": 4275, "loss": 0.1826, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.6360485113662216e-05, "epoch": 0.8888888888888888, "percentage": 17.78, "elapsed_time": "1:15:14", "remaining_time": "5:48:00"}
+{"current_steps": 770, "total_steps": 4275, "loss": 0.155, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.626399771610739e-05, "epoch": 0.9005847953216374, "percentage": 18.01, "elapsed_time": "1:16:05", "remaining_time": "5:46:21"}
+{"current_steps": 780, "total_steps": 4275, "loss": 0.1791, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.616635115808705e-05, "epoch": 0.9122807017543859, "percentage": 18.25, "elapsed_time": "1:16:58", "remaining_time": "5:44:52"}
+{"current_steps": 790, "total_steps": 4275, "loss": 0.167, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.606755076259039e-05, "epoch": 0.9239766081871345, "percentage": 18.48, "elapsed_time": "1:17:42", "remaining_time": "5:42:47"}
+{"current_steps": 800, "total_steps": 4275, "loss": 0.1663, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.5967601915505505e-05, "epoch": 0.935672514619883, "percentage": 18.71, "elapsed_time": "1:18:33", "remaining_time": "5:41:16"}
+{"current_steps": 800, "total_steps": 4275, "loss": null, "eval_loss": 0.1684267818927765, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 0.935672514619883, "percentage": 18.71, "elapsed_time": "1:18:33", "remaining_time": "5:41:16"}
+{"current_steps": 810, "total_steps": 4275, "loss": 0.1783, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.5866510065325837e-05, "epoch": 0.9473684210526315, "percentage": 18.95, "elapsed_time": "1:20:35", "remaining_time": "5:44:46"}
+{"current_steps": 820, "total_steps": 4275, "loss": 0.1765, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.576428072285316e-05, "epoch": 0.9590643274853801, "percentage": 19.18, "elapsed_time": "1:21:25", "remaining_time": "5:43:06"}
+{"current_steps": 830, "total_steps": 4275, "loss": 0.1485, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.5660919460897136e-05, "epoch": 0.9707602339181286, "percentage": 19.42, "elapsed_time": "1:22:22", "remaining_time": "5:41:55"}
+{"current_steps": 840, "total_steps": 4275, "loss": 0.1719, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.5556431913971564e-05, "epoch": 0.9824561403508771, "percentage": 19.65, "elapsed_time": "1:23:20", "remaining_time": "5:40:48"}
+{"current_steps": 850, "total_steps": 4275, "loss": 0.1671, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.5450823777987224e-05, "epoch": 0.9941520467836257, "percentage": 19.88, "elapsed_time": "1:24:15", "remaining_time": "5:39:32"}
+{"current_steps": 860, "total_steps": 4275, "loss": 0.1546, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.534410080994134e-05, "epoch": 1.0058479532163742, "percentage": 20.12, "elapsed_time": "1:25:08", "remaining_time": "5:38:05"}
+{"current_steps": 870, "total_steps": 4275, "loss": 0.1423, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.523626882760379e-05, "epoch": 1.0175438596491229, "percentage": 20.35, "elapsed_time": "1:26:04", "remaining_time": "5:36:53"}
+{"current_steps": 880, "total_steps": 4275, "loss": 0.1596, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.512733370919994e-05, "epoch": 1.0292397660818713, "percentage": 20.58, "elapsed_time": "1:26:52", "remaining_time": "5:35:09"}
+{"current_steps": 890, "total_steps": 4275, "loss": 0.1554, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.5017301393090226e-05, "epoch": 1.04093567251462, "percentage": 20.82, "elapsed_time": "1:27:52", "remaining_time": "5:34:12"}
+{"current_steps": 900, "total_steps": 4275, "loss": 0.1553, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.490617787744641e-05, "epoch": 1.0526315789473684, "percentage": 21.05, "elapsed_time": "1:28:43", "remaining_time": "5:32:42"}
+{"current_steps": 900, "total_steps": 4275, "loss": null, "eval_loss": 0.15236565470695496, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 1.0526315789473684, "percentage": 21.05, "elapsed_time": "1:28:43", "remaining_time": "5:32:42"}
+{"current_steps": 910, "total_steps": 4275, "loss": 0.1656, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.479396921992462e-05, "epoch": 1.064327485380117, "percentage": 21.29, "elapsed_time": "1:30:51", "remaining_time": "5:35:57"}
+{"current_steps": 920, "total_steps": 4275, "loss": 0.1604, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.4680681537335125e-05, "epoch": 1.0760233918128654, "percentage": 21.52, "elapsed_time": "1:31:47", "remaining_time": "5:34:44"}
+{"current_steps": 930, "total_steps": 4275, "loss": 0.1695, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.456632100530891e-05, "epoch": 1.087719298245614, "percentage": 21.75, "elapsed_time": "1:32:38", "remaining_time": "5:33:12"}
+{"current_steps": 940, "total_steps": 4275, "loss": 0.1528, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.445089385796099e-05, "epoch": 1.0994152046783625, "percentage": 21.99, "elapsed_time": "1:33:29", "remaining_time": "5:31:43"}
+{"current_steps": 950, "total_steps": 4275, "loss": 0.1409, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.4334406387550584e-05, "epoch": 1.1111111111111112, "percentage": 22.22, "elapsed_time": "1:34:25", "remaining_time": "5:30:30"}
+{"current_steps": 960, "total_steps": 4275, "loss": 0.151, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.4216864944138134e-05, "epoch": 1.1228070175438596, "percentage": 22.46, "elapsed_time": "1:35:10", "remaining_time": "5:28:40"}
+{"current_steps": 970, "total_steps": 4275, "loss": 0.1443, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.4098275935239105e-05, "epoch": 1.1345029239766082, "percentage": 22.69, "elapsed_time": "1:36:05", "remaining_time": "5:27:22"}
+{"current_steps": 980, "total_steps": 4275, "loss": 0.1723, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.39786458254747e-05, "epoch": 1.1461988304093567, "percentage": 22.92, "elapsed_time": "1:36:58", "remaining_time": "5:26:03"}
+{"current_steps": 990, "total_steps": 4275, "loss": 0.1642, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.385798113621949e-05, "epoch": 1.1578947368421053, "percentage": 23.16, "elapsed_time": "1:37:54", "remaining_time": "5:24:54"}
+{"current_steps": 1000, "total_steps": 4275, "loss": 0.1665, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.373628844524587e-05, "epoch": 1.1695906432748537, "percentage": 23.39, "elapsed_time": "1:38:53", "remaining_time": "5:23:51"}
+{"current_steps": 1000, "total_steps": 4275, "loss": null, "eval_loss": 0.15835003554821014, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 1.1695906432748537, "percentage": 23.39, "elapsed_time": "1:38:53", "remaining_time": "5:23:51"}
+{"current_steps": 1010, "total_steps": 4275, "loss": 0.149, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.361357438636552e-05, "epoch": 1.1812865497076024, "percentage": 23.63, "elapsed_time": "1:40:48", "remaining_time": "5:25:53"}
+{"current_steps": 1020, "total_steps": 4275, "loss": 0.1487, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.3489845649067756e-05, "epoch": 1.1929824561403508, "percentage": 23.86, "elapsed_time": "1:41:32", "remaining_time": "5:24:00"}
+{"current_steps": 1030, "total_steps": 4275, "loss": 0.1722, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.336510897815489e-05, "epoch": 1.2046783625730995, "percentage": 24.09, "elapsed_time": "1:42:21", "remaining_time": "5:22:27"}
+{"current_steps": 1040, "total_steps": 4275, "loss": 0.1586, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.323937117337451e-05, "epoch": 1.2163742690058479, "percentage": 24.33, "elapsed_time": "1:43:16", "remaining_time": "5:21:13"}
+{"current_steps": 1050, "total_steps": 4275, "loss": 0.1372, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.3112639089048856e-05, "epoch": 1.2280701754385965, "percentage": 24.56, "elapsed_time": "1:44:17", "remaining_time": "5:20:18"}
+{"current_steps": 1060, "total_steps": 4275, "loss": 0.1375, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.298491963370115e-05, "epoch": 1.239766081871345, "percentage": 24.8, "elapsed_time": "1:45:11", "remaining_time": "5:19:03"}
+{"current_steps": 1070, "total_steps": 4275, "loss": 0.1429, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.285621976967898e-05, "epoch": 1.2514619883040936, "percentage": 25.03, "elapsed_time": "1:45:56", "remaining_time": "5:17:19"}
+{"current_steps": 1080, "total_steps": 4275, "loss": 0.1464, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.272654651277478e-05, "epoch": 1.263157894736842, "percentage": 25.26, "elapsed_time": "1:46:45", "remaining_time": "5:15:49"}
+{"current_steps": 1090, "total_steps": 4275, "loss": 0.143, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.259590693184337e-05, "epoch": 1.2748538011695907, "percentage": 25.5, "elapsed_time": "1:47:29", "remaining_time": "5:14:04"}
+{"current_steps": 1100, "total_steps": 4275, "loss": 0.1424, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.2464308148416626e-05, "epoch": 1.286549707602339, "percentage": 25.73, "elapsed_time": "1:48:29", "remaining_time": "5:13:07"}
+{"current_steps": 1100, "total_steps": 4275, "loss": null, "eval_loss": 0.16600148379802704, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 1.286549707602339, "percentage": 25.73, "elapsed_time": "1:48:29", "remaining_time": "5:13:07"}
+{"current_steps": 1110, "total_steps": 4275, "loss": 0.147, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.233175733631525e-05, "epoch": 1.2982456140350878, "percentage": 25.96, "elapsed_time": "1:50:36", "remaining_time": "5:15:22"}
+{"current_steps": 1120, "total_steps": 4275, "loss": 0.1373, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.2198261721257715e-05, "epoch": 1.3099415204678362, "percentage": 26.2, "elapsed_time": "1:51:26", "remaining_time": "5:13:56"}
+{"current_steps": 1130, "total_steps": 4275, "loss": 0.1439, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.206382858046636e-05, "epoch": 1.3216374269005848, "percentage": 26.43, "elapsed_time": "1:52:24", "remaining_time": "5:12:50"}
+{"current_steps": 1140, "total_steps": 4275, "loss": 0.1372, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.1928465242270696e-05, "epoch": 1.3333333333333333, "percentage": 26.67, "elapsed_time": "1:53:17", "remaining_time": "5:11:34"}
+{"current_steps": 1150, "total_steps": 4275, "loss": 0.1561, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.1792179085707935e-05, "epoch": 1.345029239766082, "percentage": 26.9, "elapsed_time": "1:54:11", "remaining_time": "5:10:18"}
+{"current_steps": 1160, "total_steps": 4275, "loss": 0.1354, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.1654977540120686e-05, "epoch": 1.3567251461988303, "percentage": 27.13, "elapsed_time": "1:55:04", "remaining_time": "5:08:59"}
+{"current_steps": 1170, "total_steps": 4275, "loss": 0.1372, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.151686808475204e-05, "epoch": 1.368421052631579, "percentage": 27.37, "elapsed_time": "1:55:52", "remaining_time": "5:07:30"}
+{"current_steps": 1180, "total_steps": 4275, "loss": 0.1776, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.137785824833776e-05, "epoch": 1.3801169590643274, "percentage": 27.6, "elapsed_time": "1:56:51", "remaining_time": "5:06:31"}
+{"current_steps": 1190, "total_steps": 4275, "loss": 0.1475, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.123795560869595e-05, "epoch": 1.391812865497076, "percentage": 27.84, "elapsed_time": "1:57:45", "remaining_time": "5:05:15"}
+{"current_steps": 1200, "total_steps": 4275, "loss": 0.1163, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.109716779231393e-05, "epoch": 1.4035087719298245, "percentage": 28.07, "elapsed_time": "1:58:32", "remaining_time": "5:03:46"}
+{"current_steps": 1200, "total_steps": 4275, "loss": null, "eval_loss": 0.14416372776031494, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 1.4035087719298245, "percentage": 28.07, "elapsed_time": "1:58:32", "remaining_time": "5:03:46"}
+{"current_steps": 1210, "total_steps": 4275, "loss": 0.1398, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.0955502473932495e-05, "epoch": 1.4152046783625731, "percentage": 28.3, "elapsed_time": "2:00:36", "remaining_time": "5:05:30"}
+{"current_steps": 1220, "total_steps": 4275, "loss": 0.1424, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.0812967376127555e-05, "epoch": 1.4269005847953216, "percentage": 28.54, "elapsed_time": "2:01:32", "remaining_time": "5:04:20"}
+{"current_steps": 1230, "total_steps": 4275, "loss": 0.125, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.0669570268889124e-05, "epoch": 1.4385964912280702, "percentage": 28.77, "elapsed_time": "2:02:24", "remaining_time": "5:03:03"}
+{"current_steps": 1240, "total_steps": 4275, "loss": 0.1474, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.0525318969197825e-05, "epoch": 1.4502923976608186, "percentage": 29.01, "elapsed_time": "2:03:10", "remaining_time": "5:01:29"}
+{"current_steps": 1250, "total_steps": 4275, "loss": 0.1405, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.038022134059868e-05, "epoch": 1.4619883040935673, "percentage": 29.24, "elapsed_time": "2:04:06", "remaining_time": "5:00:21"}
+{"current_steps": 1260, "total_steps": 4275, "loss": 0.1226, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.02342852927725e-05, "epoch": 1.4736842105263157, "percentage": 29.47, "elapsed_time": "2:05:01", "remaining_time": "4:59:10"}
+{"current_steps": 1270, "total_steps": 4275, "loss": 0.1362, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.00875187811047e-05, "epoch": 1.4853801169590644, "percentage": 29.71, "elapsed_time": "2:05:55", "remaining_time": "4:57:58"}
+{"current_steps": 1280, "total_steps": 4275, "loss": 0.1541, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.993992980625161e-05, "epoch": 1.4970760233918128, "percentage": 29.94, "elapsed_time": "2:06:45", "remaining_time": "4:56:34"}
+{"current_steps": 1290, "total_steps": 4275, "loss": 0.1429, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.979152641370434e-05, "epoch": 1.5087719298245614, "percentage": 30.18, "elapsed_time": "2:07:37", "remaining_time": "4:55:20"}
+{"current_steps": 1300, "total_steps": 4275, "loss": 0.1593, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.964231669335023e-05, "epoch": 1.52046783625731, "percentage": 30.41, "elapsed_time": "2:08:30", "remaining_time": "4:54:05"}
+{"current_steps": 1300, "total_steps": 4275, "loss": null, "eval_loss": 0.136515274643898, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 1.52046783625731, "percentage": 30.41, "elapsed_time": "2:08:30", "remaining_time": "4:54:05"}
+{"current_steps": 1310, "total_steps": 4275, "loss": 0.1437, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.94923087790318e-05, "epoch": 1.5321637426900585, "percentage": 30.64, "elapsed_time": "2:10:32", "remaining_time": "4:55:28"}
+{"current_steps": 1320, "total_steps": 4275, "loss": 0.135, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.934151084810338e-05, "epoch": 1.543859649122807, "percentage": 30.88, "elapsed_time": "2:11:13", "remaining_time": "4:53:45"}
+{"current_steps": 1330, "total_steps": 4275, "loss": 0.1278, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.918993112098532e-05, "epoch": 1.5555555555555556, "percentage": 31.11, "elapsed_time": "2:12:08", "remaining_time": "4:52:35"}
+{"current_steps": 1340, "total_steps": 4275, "loss": 0.1384, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.9037577860715915e-05, "epoch": 1.5672514619883042, "percentage": 31.35, "elapsed_time": "2:13:05", "remaining_time": "4:51:31"}
+{"current_steps": 1350, "total_steps": 4275, "loss": 0.1262, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.888445937250091e-05, "epoch": 1.5789473684210527, "percentage": 31.58, "elapsed_time": "2:13:59", "remaining_time": "4:50:19"}
+{"current_steps": 1360, "total_steps": 4275, "loss": 0.1419, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.873058400326078e-05, "epoch": 1.590643274853801, "percentage": 31.81, "elapsed_time": "2:14:51", "remaining_time": "4:49:03"}
+{"current_steps": 1370, "total_steps": 4275, "loss": 0.1429, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.8575960141175735e-05, "epoch": 1.6023391812865497, "percentage": 32.05, "elapsed_time": "2:15:45", "remaining_time": "4:47:51"}
+{"current_steps": 1380, "total_steps": 4275, "loss": 0.1402, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.842059621522843e-05, "epoch": 1.6140350877192984, "percentage": 32.28, "elapsed_time": "2:16:33", "remaining_time": "4:46:27"}
+{"current_steps": 1390, "total_steps": 4275, "loss": 0.1295, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.826450069474447e-05, "epoch": 1.6257309941520468, "percentage": 32.51, "elapsed_time": "2:17:29", "remaining_time": "4:45:22"}
+{"current_steps": 1400, "total_steps": 4275, "loss": 0.1529, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.8107682088930794e-05, "epoch": 1.6374269005847952, "percentage": 32.75, "elapsed_time": "2:18:24", "remaining_time": "4:44:13"}
+{"current_steps": 1400, "total_steps": 4275, "loss": null, "eval_loss": 0.14436422288417816, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 1.6374269005847952, "percentage": 32.75, "elapsed_time": "2:18:24", "remaining_time": "4:44:13"}
+{"current_steps": 1410, "total_steps": 4275, "loss": 0.1212, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.7950148946411703e-05, "epoch": 1.6491228070175439, "percentage": 32.98, "elapsed_time": "2:20:19", "remaining_time": "4:45:07"}
+{"current_steps": 1420, "total_steps": 4275, "loss": 0.1284, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.779190985476294e-05, "epoch": 1.6608187134502925, "percentage": 33.22, "elapsed_time": "2:21:09", "remaining_time": "4:43:48"}
+{"current_steps": 1430, "total_steps": 4275, "loss": 0.1386, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.763297344004352e-05, "epoch": 1.672514619883041, "percentage": 33.45, "elapsed_time": "2:21:55", "remaining_time": "4:42:22"}
+{"current_steps": 1440, "total_steps": 4275, "loss": 0.1426, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.747334836632549e-05, "epoch": 1.6842105263157894, "percentage": 33.68, "elapsed_time": "2:22:46", "remaining_time": "4:41:05"}
+{"current_steps": 1450, "total_steps": 4275, "loss": 0.1362, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.7313043335221655e-05, "epoch": 1.695906432748538, "percentage": 33.92, "elapsed_time": "2:23:36", "remaining_time": "4:39:48"}
+{"current_steps": 1460, "total_steps": 4275, "loss": 0.1498, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.715206708541122e-05, "epoch": 1.7076023391812867, "percentage": 34.15, "elapsed_time": "2:24:31", "remaining_time": "4:38:38"}
+{"current_steps": 1470, "total_steps": 4275, "loss": 0.1371, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.69904283921634e-05, "epoch": 1.719298245614035, "percentage": 34.39, "elapsed_time": "2:25:28", "remaining_time": "4:37:34"}
+{"current_steps": 1480, "total_steps": 4275, "loss": 0.1343, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.682813606685908e-05, "epoch": 1.7309941520467835, "percentage": 34.62, "elapsed_time": "2:26:15", "remaining_time": "4:36:12"}
+{"current_steps": 1490, "total_steps": 4275, "loss": 0.1344, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.666519895651046e-05, "epoch": 1.7426900584795322, "percentage": 34.85, "elapsed_time": "2:27:07", "remaining_time": "4:34:59"}
+{"current_steps": 1500, "total_steps": 4275, "loss": 0.1321, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.6501625943278805e-05, "epoch": 1.7543859649122808, "percentage": 35.09, "elapsed_time": "2:27:56", "remaining_time": "4:33:42"}
+{"current_steps": 1500, "total_steps": 4275, "loss": null, "eval_loss": 0.13336236774921417, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 1.7543859649122808, "percentage": 35.09, "elapsed_time": "2:27:56", "remaining_time": "4:33:42"}
+{"current_steps": 1510, "total_steps": 4275, "loss": 0.1102, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.6337425943990224e-05, "epoch": 1.7660818713450293, "percentage": 35.32, "elapsed_time": "2:29:55", "remaining_time": "4:34:32"}
+{"current_steps": 1520, "total_steps": 4275, "loss": 0.1352, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.6172607909649606e-05, "epoch": 1.7777777777777777, "percentage": 35.56, "elapsed_time": "2:30:47", "remaining_time": "4:33:18"}
+{"current_steps": 1530, "total_steps": 4275, "loss": 0.1296, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.600718082495269e-05, "epoch": 1.7894736842105263, "percentage": 35.79, "elapsed_time": "2:31:43", "remaining_time": "4:32:13"}
+{"current_steps": 1540, "total_steps": 4275, "loss": 0.1427, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.5841153707796245e-05, "epoch": 1.801169590643275, "percentage": 36.02, "elapsed_time": "2:32:33", "remaining_time": "4:30:56"}
+{"current_steps": 1550, "total_steps": 4275, "loss": 0.1426, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.567453560878652e-05, "epoch": 1.8128654970760234, "percentage": 36.26, "elapsed_time": "2:33:30", "remaining_time": "4:29:53"}
+{"current_steps": 1560, "total_steps": 4275, "loss": 0.1342, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.550733561074582e-05, "epoch": 1.8245614035087718, "percentage": 36.49, "elapsed_time": "2:34:26", "remaining_time": "4:28:47"}
+{"current_steps": 1570, "total_steps": 4275, "loss": 0.1337, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.533956282821746e-05, "epoch": 1.8362573099415205, "percentage": 36.73, "elapsed_time": "2:35:20", "remaining_time": "4:27:38"}
+{"current_steps": 1580, "total_steps": 4275, "loss": 0.1306, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.51712264069688e-05, "epoch": 1.8479532163742691, "percentage": 36.96, "elapsed_time": "2:36:10", "remaining_time": "4:26:23"}
+{"current_steps": 1590, "total_steps": 4275, "loss": 0.1299, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.500233552349276e-05, "epoch": 1.8596491228070176, "percentage": 37.19, "elapsed_time": "2:37:03", "remaining_time": "4:25:13"}
+{"current_steps": 1600, "total_steps": 4275, "loss": 0.135, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.483289938450755e-05, "epoch": 1.871345029239766, "percentage": 37.43, "elapsed_time": "2:37:53", "remaining_time": "4:23:59"}
+{"current_steps": 1600, "total_steps": 4275, "loss": null, "eval_loss": 0.12928339838981628, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 1.871345029239766, "percentage": 37.43, "elapsed_time": "2:37:53", "remaining_time": "4:23:59"}
+{"current_steps": 1610, "total_steps": 4275, "loss": 0.1304, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.466292722645481e-05, "epoch": 1.8830409356725146, "percentage": 37.66, "elapsed_time": "2:39:55", "remaining_time": "4:24:43"}
+{"current_steps": 1620, "total_steps": 4275, "loss": 0.1203, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.449242831499608e-05, "epoch": 1.8947368421052633, "percentage": 37.89, "elapsed_time": "2:40:50", "remaining_time": "4:23:35"}
+{"current_steps": 1630, "total_steps": 4275, "loss": 0.1323, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.432141194450772e-05, "epoch": 1.9064327485380117, "percentage": 38.13, "elapsed_time": "2:41:35", "remaining_time": "4:22:13"}
+{"current_steps": 1640, "total_steps": 4275, "loss": 0.1159, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.4149887437574226e-05, "epoch": 1.9181286549707601, "percentage": 38.36, "elapsed_time": "2:42:28", "remaining_time": "4:21:03"}
+{"current_steps": 1650, "total_steps": 4275, "loss": 0.1272, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.397786414448004e-05, "epoch": 1.9298245614035088, "percentage": 38.6, "elapsed_time": "2:43:23", "remaining_time": "4:19:55"}
+{"current_steps": 1660, "total_steps": 4275, "loss": 0.124, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.3805351442699885e-05, "epoch": 1.9415204678362574, "percentage": 38.83, "elapsed_time": "2:44:21", "remaining_time": "4:18:54"}
+{"current_steps": 1670, "total_steps": 4275, "loss": 0.1259, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.363235873638749e-05, "epoch": 1.9532163742690059, "percentage": 39.06, "elapsed_time": "2:45:17", "remaining_time": "4:17:49"}
+{"current_steps": 1680, "total_steps": 4275, "loss": 0.1223, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.345889545586301e-05, "epoch": 1.9649122807017543, "percentage": 39.3, "elapsed_time": "2:46:11", "remaining_time": "4:16:41"}
+{"current_steps": 1690, "total_steps": 4275, "loss": 0.1312, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.328497105709888e-05, "epoch": 1.976608187134503, "percentage": 39.53, "elapsed_time": "2:47:03", "remaining_time": "4:15:31"}
+{"current_steps": 1700, "total_steps": 4275, "loss": 0.1322, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.311059502120446e-05, "epoch": 1.9883040935672516, "percentage": 39.77, "elapsed_time": "2:47:58", "remaining_time": "4:14:25"}
+{"current_steps": 1700, "total_steps": 4275, "loss": null, "eval_loss": 0.14009280502796173, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 1.9883040935672516, "percentage": 39.77, "elapsed_time": "2:47:58", "remaining_time": "4:14:25"}
+{"current_steps": 1710, "total_steps": 4275, "loss": 0.1407, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.293577685390907e-05, "epoch": 2.0, "percentage": 40.0, "elapsed_time": "2:50:00", "remaining_time": "4:15:00"}
+{"current_steps": 1720, "total_steps": 4275, "loss": 0.1053, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.2760526085043894e-05, "epoch": 2.0116959064327484, "percentage": 40.23, "elapsed_time": "2:50:43", "remaining_time": "4:13:36"}
+{"current_steps": 1730, "total_steps": 4275, "loss": 0.1356, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.258485226802242e-05, "epoch": 2.023391812865497, "percentage": 40.47, "elapsed_time": "2:51:44", "remaining_time": "4:12:39"}
+{"current_steps": 1740, "total_steps": 4275, "loss": 0.1009, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.2408764979319695e-05, "epoch": 2.0350877192982457, "percentage": 40.7, "elapsed_time": "2:52:38", "remaining_time": "4:11:30"}
+{"current_steps": 1750, "total_steps": 4275, "loss": 0.1134, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.2232273817950273e-05, "epoch": 2.046783625730994, "percentage": 40.94, "elapsed_time": "2:53:30", "remaining_time": "4:10:20"}
+{"current_steps": 1760, "total_steps": 4275, "loss": 0.1108, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.205538840494494e-05, "epoch": 2.0584795321637426, "percentage": 41.17, "elapsed_time": "2:54:28", "remaining_time": "4:09:19"}
+{"current_steps": 1770, "total_steps": 4275, "loss": 0.1126, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.187811838282627e-05, "epoch": 2.0701754385964914, "percentage": 41.4, "elapsed_time": "2:55:26", "remaining_time": "4:08:17"}
+{"current_steps": 1780, "total_steps": 4275, "loss": 0.1284, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.170047341508293e-05, "epoch": 2.08187134502924, "percentage": 41.64, "elapsed_time": "2:56:14", "remaining_time": "4:07:02"}
+{"current_steps": 1790, "total_steps": 4275, "loss": 0.1202, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.1522463185642965e-05, "epoch": 2.0935672514619883, "percentage": 41.87, "elapsed_time": "2:57:08", "remaining_time": "4:05:55"}
+{"current_steps": 1800, "total_steps": 4275, "loss": 0.1355, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.134409739834584e-05, "epoch": 2.1052631578947367, "percentage": 42.11, "elapsed_time": "2:58:02", "remaining_time": "4:04:48"}
+{"current_steps": 1800, "total_steps": 4275, "loss": null, "eval_loss": 0.12916673719882965, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 2.1052631578947367, "percentage": 42.11, "elapsed_time": "2:58:02", "remaining_time": "4:04:48"}
+{"current_steps": 1810, "total_steps": 4275, "loss": 0.1225, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.11653857764135e-05, "epoch": 2.116959064327485, "percentage": 42.34, "elapsed_time": "3:00:10", "remaining_time": "4:05:22"}
+{"current_steps": 1820, "total_steps": 4275, "loss": 0.1103, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.098633806192029e-05, "epoch": 2.128654970760234, "percentage": 42.57, "elapsed_time": "3:01:07", "remaining_time": "4:04:19"}
+{"current_steps": 1830, "total_steps": 4275, "loss": 0.1042, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.080696401526193e-05, "epoch": 2.1403508771929824, "percentage": 42.81, "elapsed_time": "3:02:04", "remaining_time": "4:03:16"}
+{"current_steps": 1840, "total_steps": 4275, "loss": 0.1336, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.062727341462342e-05, "epoch": 2.152046783625731, "percentage": 43.04, "elapsed_time": "3:03:05", "remaining_time": "4:02:17"}
+{"current_steps": 1850, "total_steps": 4275, "loss": 0.1212, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.0447276055445984e-05, "epoch": 2.1637426900584797, "percentage": 43.27, "elapsed_time": "3:04:00", "remaining_time": "4:01:12"}
+{"current_steps": 1860, "total_steps": 4275, "loss": 0.1079, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.0266981749893157e-05, "epoch": 2.175438596491228, "percentage": 43.51, "elapsed_time": "3:04:58", "remaining_time": "4:00:10"}
+{"current_steps": 1870, "total_steps": 4275, "loss": 0.1505, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.008640032631585e-05, "epoch": 2.1871345029239766, "percentage": 43.74, "elapsed_time": "3:05:45", "remaining_time": "3:58:54"}
+{"current_steps": 1880, "total_steps": 4275, "loss": 0.1306, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.990554162871656e-05, "epoch": 2.198830409356725, "percentage": 43.98, "elapsed_time": "3:06:44", "remaining_time": "3:57:53"}
+{"current_steps": 1890, "total_steps": 4275, "loss": 0.1184, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.97244155162128e-05, "epoch": 2.2105263157894735, "percentage": 44.21, "elapsed_time": "3:07:34", "remaining_time": "3:56:41"}
+{"current_steps": 1900, "total_steps": 4275, "loss": 0.1227, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.954303186249962e-05, "epoch": 2.2222222222222223, "percentage": 44.44, "elapsed_time": "3:08:20", "remaining_time": "3:55:25"}
+{"current_steps": 1900, "total_steps": 4275, "loss": null, "eval_loss": 0.13539400696754456, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 2.2222222222222223, "percentage": 44.44, "elapsed_time": "3:08:20", "remaining_time": "3:55:25"}
+{"current_steps": 1910, "total_steps": 4275, "loss": 0.1208, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.936140055531135e-05, "epoch": 2.2339181286549707, "percentage": 44.68, "elapsed_time": "3:10:22", "remaining_time": "3:55:44"}
+{"current_steps": 1920, "total_steps": 4275, "loss": 0.1231, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.9179531495882644e-05, "epoch": 2.245614035087719, "percentage": 44.91, "elapsed_time": "3:11:12", "remaining_time": "3:54:31"}
+{"current_steps": 1930, "total_steps": 4275, "loss": 0.123, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.8997434598408653e-05, "epoch": 2.257309941520468, "percentage": 45.15, "elapsed_time": "3:12:07", "remaining_time": "3:53:26"}
+{"current_steps": 1940, "total_steps": 4275, "loss": 0.1132, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.8815119789504658e-05, "epoch": 2.2690058479532165, "percentage": 45.38, "elapsed_time": "3:12:59", "remaining_time": "3:52:17"}
+{"current_steps": 1950, "total_steps": 4275, "loss": 0.1049, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.8632597007664895e-05, "epoch": 2.280701754385965, "percentage": 45.61, "elapsed_time": "3:13:53", "remaining_time": "3:51:10"}
+{"current_steps": 1960, "total_steps": 4275, "loss": 0.111, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.8449876202720793e-05, "epoch": 2.2923976608187133, "percentage": 45.85, "elapsed_time": "3:14:46", "remaining_time": "3:50:03"}
+{"current_steps": 1970, "total_steps": 4275, "loss": 0.1012, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.8266967335298584e-05, "epoch": 2.3040935672514617, "percentage": 46.08, "elapsed_time": "3:15:33", "remaining_time": "3:48:49"}
+{"current_steps": 1980, "total_steps": 4275, "loss": 0.1121, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.8083880376276307e-05, "epoch": 2.3157894736842106, "percentage": 46.32, "elapsed_time": "3:16:24", "remaining_time": "3:47:38"}
+{"current_steps": 1990, "total_steps": 4275, "loss": 0.1222, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.7900625306240263e-05, "epoch": 2.327485380116959, "percentage": 46.55, "elapsed_time": "3:17:20", "remaining_time": "3:46:36"}
+{"current_steps": 2000, "total_steps": 4275, "loss": 0.1061, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.7717212114940988e-05, "epoch": 2.3391812865497075, "percentage": 46.78, "elapsed_time": "3:18:17", "remaining_time": "3:45:33"}
+{"current_steps": 2000, "total_steps": 4275, "loss": null, "eval_loss": 0.12341146171092987, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 2.3391812865497075, "percentage": 46.78, "elapsed_time": "3:18:17", "remaining_time": "3:45:33"}
+{"current_steps": 2010, "total_steps": 4275, "loss": 0.1144, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.7533650800748645e-05, "epoch": 2.3508771929824563, "percentage": 47.02, "elapsed_time": "3:20:18", "remaining_time": "3:45:43"}
+{"current_steps": 2020, "total_steps": 4275, "loss": 0.1077, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.7349951370107983e-05, "epoch": 2.3625730994152048, "percentage": 47.25, "elapsed_time": "3:21:11", "remaining_time": "3:44:35"}
+{"current_steps": 2030, "total_steps": 4275, "loss": 0.1088, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.7166123836992873e-05, "epoch": 2.374269005847953, "percentage": 47.49, "elapsed_time": "3:22:05", "remaining_time": "3:43:29"}
+{"current_steps": 2040, "total_steps": 4275, "loss": 0.1268, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.698217822236041e-05, "epoch": 2.3859649122807016, "percentage": 47.72, "elapsed_time": "3:22:55", "remaining_time": "3:42:18"}
+{"current_steps": 2050, "total_steps": 4275, "loss": 0.1175, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.6798124553604653e-05, "epoch": 2.39766081871345, "percentage": 47.95, "elapsed_time": "3:23:51", "remaining_time": "3:41:16"}
+{"current_steps": 2060, "total_steps": 4275, "loss": 0.1177, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.6613972864009994e-05, "epoch": 2.409356725146199, "percentage": 48.19, "elapsed_time": "3:24:40", "remaining_time": "3:40:04"}
+{"current_steps": 2070, "total_steps": 4275, "loss": 0.1343, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.642973319220422e-05, "epoch": 2.4210526315789473, "percentage": 48.42, "elapsed_time": "3:25:33", "remaining_time": "3:38:57"}
+{"current_steps": 2080, "total_steps": 4275, "loss": 0.1335, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.624541558161125e-05, "epoch": 2.4327485380116958, "percentage": 48.65, "elapsed_time": "3:26:33", "remaining_time": "3:37:58"}
+{"current_steps": 2090, "total_steps": 4275, "loss": 0.1208, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.606103007990371e-05, "epoch": 2.4444444444444446, "percentage": 48.89, "elapsed_time": "3:27:30", "remaining_time": "3:36:56"}
+{"current_steps": 2100, "total_steps": 4275, "loss": 0.1164, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.587658673845513e-05, "epoch": 2.456140350877193, "percentage": 49.12, "elapsed_time": "3:28:18", "remaining_time": "3:35:44"}
+{"current_steps": 2100, "total_steps": 4275, "loss": null, "eval_loss": 0.12629590928554535, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 2.456140350877193, "percentage": 49.12, "elapsed_time": "3:28:18", "remaining_time": "3:35:44"}
+{"current_steps": 2110, "total_steps": 4275, "loss": 0.1408, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.569209561179206e-05, "epoch": 2.4678362573099415, "percentage": 49.36, "elapsed_time": "3:30:24", "remaining_time": "3:35:53"}
+{"current_steps": 2120, "total_steps": 4275, "loss": 0.1203, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.550756675704596e-05, "epoch": 2.47953216374269, "percentage": 49.59, "elapsed_time": "3:31:21", "remaining_time": "3:34:51"}
+{"current_steps": 2130, "total_steps": 4275, "loss": 0.1187, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.5323010233404925e-05, "epoch": 2.4912280701754383, "percentage": 49.82, "elapsed_time": "3:32:19", "remaining_time": "3:33:49"}
+{"current_steps": 2140, "total_steps": 4275, "loss": 0.1105, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.5138436101565405e-05, "epoch": 2.502923976608187, "percentage": 50.06, "elapsed_time": "3:33:05", "remaining_time": "3:32:35"}
+{"current_steps": 2150, "total_steps": 4275, "loss": 0.1263, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.4953854423183694e-05, "epoch": 2.5146198830409356, "percentage": 50.29, "elapsed_time": "3:33:56", "remaining_time": "3:31:27"}
+{"current_steps": 2160, "total_steps": 4275, "loss": 0.1402, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.4769275260327474e-05, "epoch": 2.526315789473684, "percentage": 50.53, "elapsed_time": "3:34:47", "remaining_time": "3:30:19"}
+{"current_steps": 2170, "total_steps": 4275, "loss": 0.1254, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.4584708674927308e-05, "epoch": 2.538011695906433, "percentage": 50.76, "elapsed_time": "3:35:36", "remaining_time": "3:29:08"}
+{"current_steps": 2180, "total_steps": 4275, "loss": 0.1069, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.44001647282281e-05, "epoch": 2.5497076023391814, "percentage": 50.99, "elapsed_time": "3:36:20", "remaining_time": "3:27:54"}
+{"current_steps": 2190, "total_steps": 4275, "loss": 0.1203, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.4215653480240685e-05, "epoch": 2.56140350877193, "percentage": 51.23, "elapsed_time": "3:37:08", "remaining_time": "3:26:44"}
+{"current_steps": 2200, "total_steps": 4275, "loss": 0.1038, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.4031184989193384e-05, "epoch": 2.573099415204678, "percentage": 51.46, "elapsed_time": "3:38:02", "remaining_time": "3:25:39"}
+{"current_steps": 2200, "total_steps": 4275, "loss": null, "eval_loss": 0.12356504052877426, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 2.573099415204678, "percentage": 51.46, "elapsed_time": "3:38:02", "remaining_time": "3:25:39"}
+{"current_steps": 2210, "total_steps": 4275, "loss": 0.1315, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.3846769310983727e-05, "epoch": 2.5847953216374266, "percentage": 51.7, "elapsed_time": "3:40:00", "remaining_time": "3:25:33"}
+{"current_steps": 2220, "total_steps": 4275, "loss": 0.1228, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.3662416498630257e-05, "epoch": 2.5964912280701755, "percentage": 51.93, "elapsed_time": "3:40:56", "remaining_time": "3:24:31"}
+{"current_steps": 2230, "total_steps": 4275, "loss": 0.115, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.3478136601724515e-05, "epoch": 2.608187134502924, "percentage": 52.16, "elapsed_time": "3:41:56", "remaining_time": "3:23:31"}
+{"current_steps": 2240, "total_steps": 4275, "loss": 0.0981, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.329393966588323e-05, "epoch": 2.6198830409356724, "percentage": 52.4, "elapsed_time": "3:42:37", "remaining_time": "3:22:15"}
+{"current_steps": 2250, "total_steps": 4275, "loss": 0.0954, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.3109835732200686e-05, "epoch": 2.6315789473684212, "percentage": 52.63, "elapsed_time": "3:43:30", "remaining_time": "3:21:09"}
+{"current_steps": 2260, "total_steps": 4275, "loss": 0.1046, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.292583483670136e-05, "epoch": 2.6432748538011697, "percentage": 52.87, "elapsed_time": "3:44:31", "remaining_time": "3:20:11"}
+{"current_steps": 2270, "total_steps": 4275, "loss": 0.1245, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.2741947009792817e-05, "epoch": 2.654970760233918, "percentage": 53.1, "elapsed_time": "3:45:20", "remaining_time": "3:19:02"}
+{"current_steps": 2280, "total_steps": 4275, "loss": 0.1303, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.2558182275718953e-05, "epoch": 2.6666666666666665, "percentage": 53.33, "elapsed_time": "3:46:15", "remaining_time": "3:17:58"}
+{"current_steps": 2290, "total_steps": 4275, "loss": 0.1142, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.2374550652013492e-05, "epoch": 2.678362573099415, "percentage": 53.57, "elapsed_time": "3:47:09", "remaining_time": "3:16:54"}
+{"current_steps": 2300, "total_steps": 4275, "loss": 0.1196, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.219106214895397e-05, "epoch": 2.690058479532164, "percentage": 53.8, "elapsed_time": "3:48:10", "remaining_time": "3:15:55"}
+{"current_steps": 2300, "total_steps": 4275, "loss": null, "eval_loss": 0.12174443155527115, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 2.690058479532164, "percentage": 53.8, "elapsed_time": "3:48:10", "remaining_time": "3:15:55"}
+{"current_steps": 2310, "total_steps": 4275, "loss": 0.1071, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.2007726769016006e-05, "epoch": 2.7017543859649122, "percentage": 54.04, "elapsed_time": "3:50:14", "remaining_time": "3:15:51"}
+{"current_steps": 2320, "total_steps": 4275, "loss": 0.1246, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.182455450632803e-05, "epoch": 2.7134502923976607, "percentage": 54.27, "elapsed_time": "3:51:05", "remaining_time": "3:14:43"}
+{"current_steps": 2330, "total_steps": 4275, "loss": 0.1188, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.1641555346126512e-05, "epoch": 2.7251461988304095, "percentage": 54.5, "elapsed_time": "3:52:00", "remaining_time": "3:13:40"}
+{"current_steps": 2340, "total_steps": 4275, "loss": 0.101, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.1458739264211618e-05, "epoch": 2.736842105263158, "percentage": 54.74, "elapsed_time": "3:52:52", "remaining_time": "3:12:33"}
+{"current_steps": 2350, "total_steps": 4275, "loss": 0.1068, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.127611622640336e-05, "epoch": 2.7485380116959064, "percentage": 54.97, "elapsed_time": "3:53:37", "remaining_time": "3:11:22"}
+{"current_steps": 2360, "total_steps": 4275, "loss": 0.1196, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.1093696187998424e-05, "epoch": 2.760233918128655, "percentage": 55.2, "elapsed_time": "3:54:28", "remaining_time": "3:10:15"}
+{"current_steps": 2370, "total_steps": 4275, "loss": 0.1076, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.0911489093227395e-05, "epoch": 2.7719298245614032, "percentage": 55.44, "elapsed_time": "3:55:21", "remaining_time": "3:09:10"}
+{"current_steps": 2380, "total_steps": 4275, "loss": 0.111, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.0729504874712713e-05, "epoch": 2.783625730994152, "percentage": 55.67, "elapsed_time": "3:56:15", "remaining_time": "3:08:07"}
+{"current_steps": 2390, "total_steps": 4275, "loss": 0.1079, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.0547753452927197e-05, "epoch": 2.7953216374269005, "percentage": 55.91, "elapsed_time": "3:57:05", "remaining_time": "3:06:59"}
+{"current_steps": 2400, "total_steps": 4275, "loss": 0.1058, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.036624473565328e-05, "epoch": 2.807017543859649, "percentage": 56.14, "elapsed_time": "3:57:51", "remaining_time": "3:05:49"}
+{"current_steps": 2400, "total_steps": 4275, "loss": null, "eval_loss": 0.1205640360713005, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 2.807017543859649, "percentage": 56.14, "elapsed_time": "3:57:51", "remaining_time": "3:05:49"}
+{"current_steps": 2410, "total_steps": 4275, "loss": 0.1111, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.0184988617442852e-05, "epoch": 2.818713450292398, "percentage": 56.37, "elapsed_time": "3:59:59", "remaining_time": "3:05:43"}
+{"current_steps": 2420, "total_steps": 4275, "loss": 0.1069, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.0003994979077937e-05, "epoch": 2.8304093567251463, "percentage": 56.61, "elapsed_time": "4:00:52", "remaining_time": "3:04:38"}
+{"current_steps": 2430, "total_steps": 4275, "loss": 0.1153, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.9823273687032045e-05, "epoch": 2.8421052631578947, "percentage": 56.84, "elapsed_time": "4:01:41", "remaining_time": "3:03:30"}
+{"current_steps": 2440, "total_steps": 4275, "loss": 0.1097, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.964283459293232e-05, "epoch": 2.853801169590643, "percentage": 57.08, "elapsed_time": "4:02:32", "remaining_time": "3:02:24"}
+{"current_steps": 2450, "total_steps": 4275, "loss": 0.1123, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.94626875330225e-05, "epoch": 2.8654970760233915, "percentage": 57.31, "elapsed_time": "4:03:35", "remaining_time": "3:01:27"}
+{"current_steps": 2460, "total_steps": 4275, "loss": 0.1194, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.92828423276267e-05, "epoch": 2.8771929824561404, "percentage": 57.54, "elapsed_time": "4:04:18", "remaining_time": "3:00:15"}
+{"current_steps": 2470, "total_steps": 4275, "loss": 0.1041, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.9103308780614118e-05, "epoch": 2.888888888888889, "percentage": 57.78, "elapsed_time": "4:05:16", "remaining_time": "2:59:14"}
+{"current_steps": 2480, "total_steps": 4275, "loss": 0.1155, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.8924096678864557e-05, "epoch": 2.9005847953216373, "percentage": 58.01, "elapsed_time": "4:06:06", "remaining_time": "2:58:08"}
+{"current_steps": 2490, "total_steps": 4275, "loss": 0.1073, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.8745215791734942e-05, "epoch": 2.912280701754386, "percentage": 58.25, "elapsed_time": "4:06:54", "remaining_time": "2:56:59"}
+{"current_steps": 2500, "total_steps": 4275, "loss": 0.1177, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.8566675870526756e-05, "epoch": 2.9239766081871346, "percentage": 58.48, "elapsed_time": "4:07:48", "remaining_time": "2:55:56"}
+{"current_steps": 2500, "total_steps": 4275, "loss": null, "eval_loss": 0.12042708694934845, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 2.9239766081871346, "percentage": 58.48, "elapsed_time": "4:07:48", "remaining_time": "2:55:56"}
+{"current_steps": 2510, "total_steps": 4275, "loss": 0.1073, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.838848664795447e-05, "epoch": 2.935672514619883, "percentage": 58.71, "elapsed_time": "4:09:49", "remaining_time": "2:55:40"}
+{"current_steps": 2520, "total_steps": 4275, "loss": 0.11, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.8210657837614965e-05, "epoch": 2.9473684210526314, "percentage": 58.95, "elapsed_time": "4:10:40", "remaining_time": "2:54:34"}
+{"current_steps": 2530, "total_steps": 4275, "loss": 0.112, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.803319913345805e-05, "epoch": 2.95906432748538, "percentage": 59.18, "elapsed_time": "4:11:27", "remaining_time": "2:53:26"}
+{"current_steps": 2540, "total_steps": 4275, "loss": 0.0969, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.785612020925801e-05, "epoch": 2.9707602339181287, "percentage": 59.42, "elapsed_time": "4:12:18", "remaining_time": "2:52:20"}
+{"current_steps": 2550, "total_steps": 4275, "loss": 0.1006, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.7679430718086243e-05, "epoch": 2.982456140350877, "percentage": 59.65, "elapsed_time": "4:13:08", "remaining_time": "2:51:14"}
+{"current_steps": 2560, "total_steps": 4275, "loss": 0.1041, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.7503140291785054e-05, "epoch": 2.9941520467836256, "percentage": 59.88, "elapsed_time": "4:13:56", "remaining_time": "2:50:07"}
+{"current_steps": 2570, "total_steps": 4275, "loss": 0.0973, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.732725854044261e-05, "epoch": 3.0058479532163744, "percentage": 60.12, "elapsed_time": "4:14:51", "remaining_time": "2:49:04"}
+{"current_steps": 2580, "total_steps": 4275, "loss": 0.1181, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.7151795051869025e-05, "epoch": 3.017543859649123, "percentage": 60.35, "elapsed_time": "4:15:48", "remaining_time": "2:48:03"}
+{"current_steps": 2590, "total_steps": 4275, "loss": 0.1177, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.6976759391073758e-05, "epoch": 3.0292397660818713, "percentage": 60.58, "elapsed_time": "4:16:49", "remaining_time": "2:47:05"}
+{"current_steps": 2600, "total_steps": 4275, "loss": 0.1009, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.6802161099744164e-05, "epoch": 3.0409356725146197, "percentage": 60.82, "elapsed_time": "4:17:37", "remaining_time": "2:45:58"}
+{"current_steps": 2600, "total_steps": 4275, "loss": null, "eval_loss": 0.11801114678382874, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 3.0409356725146197, "percentage": 60.82, "elapsed_time": "4:17:37", "remaining_time": "2:45:58"}
+{"current_steps": 2610, "total_steps": 4275, "loss": 0.0869, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.6628009695725346e-05, "epoch": 3.0526315789473686, "percentage": 61.05, "elapsed_time": "4:19:38", "remaining_time": "2:45:38"}
+{"current_steps": 2620, "total_steps": 4275, "loss": 0.1067, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.645431467250134e-05, "epoch": 3.064327485380117, "percentage": 61.29, "elapsed_time": "4:20:30", "remaining_time": "2:44:33"}
+{"current_steps": 2630, "total_steps": 4275, "loss": 0.1065, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.628108549867754e-05, "epoch": 3.0760233918128654, "percentage": 61.52, "elapsed_time": "4:21:24", "remaining_time": "2:43:30"}
+{"current_steps": 2640, "total_steps": 4275, "loss": 0.1046, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.610833161746461e-05, "epoch": 3.087719298245614, "percentage": 61.75, "elapsed_time": "4:22:15", "remaining_time": "2:42:25"}
+{"current_steps": 2650, "total_steps": 4275, "loss": 0.102, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.5936062446163666e-05, "epoch": 3.0994152046783627, "percentage": 61.99, "elapsed_time": "4:23:13", "remaining_time": "2:41:24"}
+{"current_steps": 2660, "total_steps": 4275, "loss": 0.0921, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.576428737565292e-05, "epoch": 3.111111111111111, "percentage": 62.22, "elapsed_time": "4:24:03", "remaining_time": "2:40:19"}
+{"current_steps": 2670, "total_steps": 4275, "loss": 0.0966, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.5593015769875746e-05, "epoch": 3.1228070175438596, "percentage": 62.46, "elapsed_time": "4:24:51", "remaining_time": "2:39:12"}
+{"current_steps": 2680, "total_steps": 4275, "loss": 0.1179, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.5422256965330263e-05, "epoch": 3.134502923976608, "percentage": 62.69, "elapsed_time": "4:25:45", "remaining_time": "2:38:10"}
+{"current_steps": 2690, "total_steps": 4275, "loss": 0.095, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.52520202705603e-05, "epoch": 3.146198830409357, "percentage": 62.92, "elapsed_time": "4:26:36", "remaining_time": "2:37:05"}
+{"current_steps": 2700, "total_steps": 4275, "loss": 0.1024, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.508231496564806e-05, "epoch": 3.1578947368421053, "percentage": 63.16, "elapsed_time": "4:27:30", "remaining_time": "2:36:03"}
+{"current_steps": 2700, "total_steps": 4275, "loss": null, "eval_loss": 0.11845976859331131, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 3.1578947368421053, "percentage": 63.16, "elapsed_time": "4:27:30", "remaining_time": "2:36:03"}
+{"current_steps": 2710, "total_steps": 4275, "loss": 0.1048, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.4913150301708162e-05, "epoch": 3.1695906432748537, "percentage": 63.39, "elapsed_time": "4:29:40", "remaining_time": "2:35:44"}
+{"current_steps": 2720, "total_steps": 4275, "loss": 0.1227, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.4744535500383372e-05, "epoch": 3.181286549707602, "percentage": 63.63, "elapsed_time": "4:30:25", "remaining_time": "2:34:36"}
+{"current_steps": 2730, "total_steps": 4275, "loss": 0.1155, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.4576479753341882e-05, "epoch": 3.192982456140351, "percentage": 63.86, "elapsed_time": "4:31:20", "remaining_time": "2:33:33"}
+{"current_steps": 2740, "total_steps": 4275, "loss": 0.1116, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.4408992221776282e-05, "epoch": 3.2046783625730995, "percentage": 64.09, "elapsed_time": "4:32:12", "remaining_time": "2:32:29"}
+{"current_steps": 2750, "total_steps": 4275, "loss": 0.1168, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.4242082035904086e-05, "epoch": 3.216374269005848, "percentage": 64.33, "elapsed_time": "4:33:02", "remaining_time": "2:31:25"}
+{"current_steps": 2760, "total_steps": 4275, "loss": 0.0942, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.4075758294470132e-05, "epoch": 3.2280701754385963, "percentage": 64.56, "elapsed_time": "4:33:56", "remaining_time": "2:30:22"}
+{"current_steps": 2770, "total_steps": 4275, "loss": 0.103, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.3910030064250462e-05, "epoch": 3.239766081871345, "percentage": 64.8, "elapsed_time": "4:34:51", "remaining_time": "2:29:19"}
+{"current_steps": 2780, "total_steps": 4275, "loss": 0.1246, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.3744906379558165e-05, "epoch": 3.2514619883040936, "percentage": 65.03, "elapsed_time": "4:35:46", "remaining_time": "2:28:18"}
+{"current_steps": 2790, "total_steps": 4275, "loss": 0.1054, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.3580396241750823e-05, "epoch": 3.263157894736842, "percentage": 65.26, "elapsed_time": "4:36:36", "remaining_time": "2:27:13"}
+{"current_steps": 2800, "total_steps": 4275, "loss": 0.0953, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.3416508618739854e-05, "epoch": 3.2748538011695905, "percentage": 65.5, "elapsed_time": "4:37:30", "remaining_time": "2:26:11"}
+{"current_steps": 2800, "total_steps": 4275, "loss": null, "eval_loss": 0.1236434280872345, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 3.2748538011695905, "percentage": 65.5, "elapsed_time": "4:37:30", "remaining_time": "2:26:11"}
+{"current_steps": 2810, "total_steps": 4275, "loss": 0.0891, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.3253252444501635e-05, "epoch": 3.2865497076023393, "percentage": 65.73, "elapsed_time": "4:39:22", "remaining_time": "2:25:39"}
+{"current_steps": 2820, "total_steps": 4275, "loss": 0.0942, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.309063661859049e-05, "epoch": 3.2982456140350878, "percentage": 65.96, "elapsed_time": "4:40:07", "remaining_time": "2:24:31"}
+{"current_steps": 2830, "total_steps": 4275, "loss": 0.0966, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.2928670005653576e-05, "epoch": 3.309941520467836, "percentage": 66.2, "elapsed_time": "4:41:02", "remaining_time": "2:23:30"}
+{"current_steps": 2840, "total_steps": 4275, "loss": 0.0863, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.2767361434947584e-05, "epoch": 3.3216374269005846, "percentage": 66.43, "elapsed_time": "4:41:55", "remaining_time": "2:22:27"}
+{"current_steps": 2850, "total_steps": 4275, "loss": 0.0938, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.2606719699857489e-05, "epoch": 3.3333333333333335, "percentage": 66.67, "elapsed_time": "4:42:48", "remaining_time": "2:21:24"}
+{"current_steps": 2860, "total_steps": 4275, "loss": 0.1161, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.244675355741717e-05, "epoch": 3.345029239766082, "percentage": 66.9, "elapsed_time": "4:43:44", "remaining_time": "2:20:23"}
+{"current_steps": 2870, "total_steps": 4275, "loss": 0.1082, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.228747172783204e-05, "epoch": 3.3567251461988303, "percentage": 67.13, "elapsed_time": "4:44:29", "remaining_time": "2:19:16"}
+{"current_steps": 2880, "total_steps": 4275, "loss": 0.0982, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.212888289400371e-05, "epoch": 3.3684210526315788, "percentage": 67.37, "elapsed_time": "4:45:25", "remaining_time": "2:18:15"}
+{"current_steps": 2890, "total_steps": 4275, "loss": 0.1026, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.1970995701056615e-05, "epoch": 3.3801169590643276, "percentage": 67.6, "elapsed_time": "4:46:13", "remaining_time": "2:17:10"}
+{"current_steps": 2900, "total_steps": 4275, "loss": 0.0905, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 1.181381875586677e-05, "epoch": 3.391812865497076, "percentage": 67.84, "elapsed_time": "4:47:04", "remaining_time": "2:16:06"}
+{"current_steps": 2900, "total_steps": 4275, "loss": null, "eval_loss": 0.12022969871759415, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 3.391812865497076, "percentage": 67.84, "elapsed_time": "4:47:04", "remaining_time": "2:16:06"}
+{"current_steps": 2900, "total_steps": 4275, "loss": null, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 3.391812865497076, "percentage": 67.84, "elapsed_time": "4:47:04", "remaining_time": "2:16:06"}
+{"current_steps": 45, "total_steps": 45, "loss": null, "eval_loss": 0.11801114678382874, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 3.391812865497076, "percentage": 100.0, "elapsed_time": "4:49:35", "remaining_time": "0:00:00"}
diff --git a/llama2_13b_peft/paragraph_segmentation/trainer_state.json b/llama2_13b_peft/paragraph_segmentation/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..79d16c876a1b828bbe9e2edb81e96b28c646c3a8
--- /dev/null
+++ b/llama2_13b_peft/paragraph_segmentation/trainer_state.json
@@ -0,0 +1,2292 @@
+{
+ "best_metric": 0.11801114678382874,
+ "best_model_checkpoint": "ckpt/llama2_13b_other/paragraph_segmentation_no_sys/checkpoint-2600",
+ "epoch": 3.391812865497076,
+ "eval_steps": 100,
+ "global_step": 2900,
+ "is_hyper_param_search": false,
+ "is_local_process_zero": true,
+ "is_world_process_zero": true,
+ "log_history": [
+ {
+ "epoch": 0.011695906432748537,
+ "grad_norm": 0.6690150499343872,
+ "learning_rate": 2.5e-05,
+ "loss": 0.781,
+ "step": 10
+ },
+ {
+ "epoch": 0.023391812865497075,
+ "grad_norm": 0.3248962163925171,
+ "learning_rate": 5e-05,
+ "loss": 0.5723,
+ "step": 20
+ },
+ {
+ "epoch": 0.03508771929824561,
+ "grad_norm": 0.4461202621459961,
+ "learning_rate": 4.999931858975852e-05,
+ "loss": 0.378,
+ "step": 30
+ },
+ {
+ "epoch": 0.04678362573099415,
+ "grad_norm": 0.4027818739414215,
+ "learning_rate": 4.999727439617963e-05,
+ "loss": 0.2932,
+ "step": 40
+ },
+ {
+ "epoch": 0.05847953216374269,
+ "grad_norm": 0.17189592123031616,
+ "learning_rate": 4.999386753069811e-05,
+ "loss": 0.2703,
+ "step": 50
+ },
+ {
+ "epoch": 0.07017543859649122,
+ "grad_norm": 0.2007831186056137,
+ "learning_rate": 4.9989098179031804e-05,
+ "loss": 0.2844,
+ "step": 60
+ },
+ {
+ "epoch": 0.08187134502923976,
+ "grad_norm": 0.21012109518051147,
+ "learning_rate": 4.998296660117151e-05,
+ "loss": 0.249,
+ "step": 70
+ },
+ {
+ "epoch": 0.0935672514619883,
+ "grad_norm": 0.3014354109764099,
+ "learning_rate": 4.997547313136682e-05,
+ "loss": 0.2521,
+ "step": 80
+ },
+ {
+ "epoch": 0.10526315789473684,
+ "grad_norm": 0.3537119925022125,
+ "learning_rate": 4.996661817810791e-05,
+ "loss": 0.2647,
+ "step": 90
+ },
+ {
+ "epoch": 0.11695906432748537,
+ "grad_norm": 0.3879987597465515,
+ "learning_rate": 4.995640222410324e-05,
+ "loss": 0.2344,
+ "step": 100
+ },
+ {
+ "epoch": 0.11695906432748537,
+ "eval_loss": 0.23539581894874573,
+ "eval_runtime": 70.7724,
+ "eval_samples_per_second": 5.087,
+ "eval_steps_per_second": 0.636,
+ "step": 100
+ },
+ {
+ "epoch": 0.1286549707602339,
+ "grad_norm": 0.658244788646698,
+ "learning_rate": 4.994482582625327e-05,
+ "loss": 0.2274,
+ "step": 110
+ },
+ {
+ "epoch": 0.14035087719298245,
+ "grad_norm": 1.490973949432373,
+ "learning_rate": 4.9931889615620075e-05,
+ "loss": 0.2664,
+ "step": 120
+ },
+ {
+ "epoch": 0.15204678362573099,
+ "grad_norm": 0.39017337560653687,
+ "learning_rate": 4.991759429739298e-05,
+ "loss": 0.2483,
+ "step": 130
+ },
+ {
+ "epoch": 0.16374269005847952,
+ "grad_norm": 1.4655972719192505,
+ "learning_rate": 4.9901940650850074e-05,
+ "loss": 0.2269,
+ "step": 140
+ },
+ {
+ "epoch": 0.17543859649122806,
+ "grad_norm": 0.4164131283760071,
+ "learning_rate": 4.9884929529315774e-05,
+ "loss": 0.2249,
+ "step": 150
+ },
+ {
+ "epoch": 0.1871345029239766,
+ "grad_norm": 0.43179380893707275,
+ "learning_rate": 4.986656186011426e-05,
+ "loss": 0.2232,
+ "step": 160
+ },
+ {
+ "epoch": 0.19883040935672514,
+ "grad_norm": 0.9226874709129333,
+ "learning_rate": 4.984683864451898e-05,
+ "loss": 0.2274,
+ "step": 170
+ },
+ {
+ "epoch": 0.21052631578947367,
+ "grad_norm": 0.5234158039093018,
+ "learning_rate": 4.982576095769801e-05,
+ "loss": 0.2153,
+ "step": 180
+ },
+ {
+ "epoch": 0.2222222222222222,
+ "grad_norm": 0.9352672696113586,
+ "learning_rate": 4.980332994865549e-05,
+ "loss": 0.2123,
+ "step": 190
+ },
+ {
+ "epoch": 0.23391812865497075,
+ "grad_norm": 0.7342891693115234,
+ "learning_rate": 4.9779546840168964e-05,
+ "loss": 0.2247,
+ "step": 200
+ },
+ {
+ "epoch": 0.23391812865497075,
+ "eval_loss": 0.2295626550912857,
+ "eval_runtime": 70.735,
+ "eval_samples_per_second": 5.089,
+ "eval_steps_per_second": 0.636,
+ "step": 200
+ },
+ {
+ "epoch": 0.24561403508771928,
+ "grad_norm": 0.19036857783794403,
+ "learning_rate": 4.9754412928722714e-05,
+ "loss": 0.2358,
+ "step": 210
+ },
+ {
+ "epoch": 0.2573099415204678,
+ "grad_norm": 0.4506204426288605,
+ "learning_rate": 4.9727929584437136e-05,
+ "loss": 0.2187,
+ "step": 220
+ },
+ {
+ "epoch": 0.26900584795321636,
+ "grad_norm": 0.5454403162002563,
+ "learning_rate": 4.970009825099398e-05,
+ "loss": 0.2077,
+ "step": 230
+ },
+ {
+ "epoch": 0.2807017543859649,
+ "grad_norm": 0.6654325127601624,
+ "learning_rate": 4.967092044555769e-05,
+ "loss": 0.2458,
+ "step": 240
+ },
+ {
+ "epoch": 0.29239766081871343,
+ "grad_norm": 0.30863338708877563,
+ "learning_rate": 4.9640397758692715e-05,
+ "loss": 0.1873,
+ "step": 250
+ },
+ {
+ "epoch": 0.30409356725146197,
+ "grad_norm": 0.4261077642440796,
+ "learning_rate": 4.960853185427676e-05,
+ "loss": 0.1758,
+ "step": 260
+ },
+ {
+ "epoch": 0.3157894736842105,
+ "grad_norm": 1.0026551485061646,
+ "learning_rate": 4.957532446941012e-05,
+ "loss": 0.2229,
+ "step": 270
+ },
+ {
+ "epoch": 0.32748538011695905,
+ "grad_norm": 0.20305348932743073,
+ "learning_rate": 4.954077741432097e-05,
+ "loss": 0.2043,
+ "step": 280
+ },
+ {
+ "epoch": 0.3391812865497076,
+ "grad_norm": 0.20843225717544556,
+ "learning_rate": 4.950489257226667e-05,
+ "loss": 0.1949,
+ "step": 290
+ },
+ {
+ "epoch": 0.3508771929824561,
+ "grad_norm": 1.1773661375045776,
+ "learning_rate": 4.9467671899431145e-05,
+ "loss": 0.2009,
+ "step": 300
+ },
+ {
+ "epoch": 0.3508771929824561,
+ "eval_loss": 0.20195595920085907,
+ "eval_runtime": 70.7382,
+ "eval_samples_per_second": 5.089,
+ "eval_steps_per_second": 0.636,
+ "step": 300
+ },
+ {
+ "epoch": 0.36257309941520466,
+ "grad_norm": 0.6430270075798035,
+ "learning_rate": 4.942911742481819e-05,
+ "loss": 0.1841,
+ "step": 310
+ },
+ {
+ "epoch": 0.3742690058479532,
+ "grad_norm": 0.6599162220954895,
+ "learning_rate": 4.938923125014094e-05,
+ "loss": 0.2091,
+ "step": 320
+ },
+ {
+ "epoch": 0.38596491228070173,
+ "grad_norm": 0.49684128165245056,
+ "learning_rate": 4.93480155497072e-05,
+ "loss": 0.2032,
+ "step": 330
+ },
+ {
+ "epoch": 0.39766081871345027,
+ "grad_norm": 0.43949800729751587,
+ "learning_rate": 4.930547257030102e-05,
+ "loss": 0.1952,
+ "step": 340
+ },
+ {
+ "epoch": 0.4093567251461988,
+ "grad_norm": 0.18739910423755646,
+ "learning_rate": 4.926160463106015e-05,
+ "loss": 0.1926,
+ "step": 350
+ },
+ {
+ "epoch": 0.42105263157894735,
+ "grad_norm": 0.42857053875923157,
+ "learning_rate": 4.921641412334963e-05,
+ "loss": 0.2085,
+ "step": 360
+ },
+ {
+ "epoch": 0.4327485380116959,
+ "grad_norm": 0.3196578621864319,
+ "learning_rate": 4.9169903510631434e-05,
+ "loss": 0.199,
+ "step": 370
+ },
+ {
+ "epoch": 0.4444444444444444,
+ "grad_norm": 0.8997694849967957,
+ "learning_rate": 4.91220753283302e-05,
+ "loss": 0.1874,
+ "step": 380
+ },
+ {
+ "epoch": 0.45614035087719296,
+ "grad_norm": 0.6457922458648682,
+ "learning_rate": 4.907293218369499e-05,
+ "loss": 0.1775,
+ "step": 390
+ },
+ {
+ "epoch": 0.4678362573099415,
+ "grad_norm": 1.4266046285629272,
+ "learning_rate": 4.9022476755657155e-05,
+ "loss": 0.2097,
+ "step": 400
+ },
+ {
+ "epoch": 0.4678362573099415,
+ "eval_loss": 0.19560903310775757,
+ "eval_runtime": 70.7773,
+ "eval_samples_per_second": 5.086,
+ "eval_steps_per_second": 0.636,
+ "step": 400
+ },
+ {
+ "epoch": 0.47953216374269003,
+ "grad_norm": 0.8755818009376526,
+ "learning_rate": 4.8970711794684345e-05,
+ "loss": 0.1872,
+ "step": 410
+ },
+ {
+ "epoch": 0.49122807017543857,
+ "grad_norm": 0.41280242800712585,
+ "learning_rate": 4.8917640122630506e-05,
+ "loss": 0.1936,
+ "step": 420
+ },
+ {
+ "epoch": 0.5029239766081871,
+ "grad_norm": 0.4296826422214508,
+ "learning_rate": 4.886326463258212e-05,
+ "loss": 0.1933,
+ "step": 430
+ },
+ {
+ "epoch": 0.5146198830409356,
+ "grad_norm": 0.6381562948226929,
+ "learning_rate": 4.880758828870045e-05,
+ "loss": 0.1874,
+ "step": 440
+ },
+ {
+ "epoch": 0.5263157894736842,
+ "grad_norm": 0.2770218849182129,
+ "learning_rate": 4.875061412605997e-05,
+ "loss": 0.184,
+ "step": 450
+ },
+ {
+ "epoch": 0.5380116959064327,
+ "grad_norm": 0.23563164472579956,
+ "learning_rate": 4.8692345250482914e-05,
+ "loss": 0.1901,
+ "step": 460
+ },
+ {
+ "epoch": 0.5497076023391813,
+ "grad_norm": 0.3889436721801758,
+ "learning_rate": 4.863278483836996e-05,
+ "loss": 0.1803,
+ "step": 470
+ },
+ {
+ "epoch": 0.5614035087719298,
+ "grad_norm": 0.23517946898937225,
+ "learning_rate": 4.857193613652711e-05,
+ "loss": 0.1688,
+ "step": 480
+ },
+ {
+ "epoch": 0.5730994152046783,
+ "grad_norm": 1.0376458168029785,
+ "learning_rate": 4.8509802461988635e-05,
+ "loss": 0.1757,
+ "step": 490
+ },
+ {
+ "epoch": 0.5847953216374269,
+ "grad_norm": 0.3648129999637604,
+ "learning_rate": 4.844638720183632e-05,
+ "loss": 0.185,
+ "step": 500
+ },
+ {
+ "epoch": 0.5847953216374269,
+ "eval_loss": 0.1873869150876999,
+ "eval_runtime": 70.7709,
+ "eval_samples_per_second": 5.087,
+ "eval_steps_per_second": 0.636,
+ "step": 500
+ },
+ {
+ "epoch": 0.5964912280701754,
+ "grad_norm": 0.1653483510017395,
+ "learning_rate": 4.8381693813014775e-05,
+ "loss": 0.1864,
+ "step": 510
+ },
+ {
+ "epoch": 0.6081871345029239,
+ "grad_norm": 0.6757428050041199,
+ "learning_rate": 4.8315725822143024e-05,
+ "loss": 0.1723,
+ "step": 520
+ },
+ {
+ "epoch": 0.6198830409356725,
+ "grad_norm": 0.8670457601547241,
+ "learning_rate": 4.824848682532224e-05,
+ "loss": 0.161,
+ "step": 530
+ },
+ {
+ "epoch": 0.631578947368421,
+ "grad_norm": 0.7731683254241943,
+ "learning_rate": 4.817998048793969e-05,
+ "loss": 0.1908,
+ "step": 540
+ },
+ {
+ "epoch": 0.6432748538011696,
+ "grad_norm": 1.1844624280929565,
+ "learning_rate": 4.811021054446897e-05,
+ "loss": 0.1888,
+ "step": 550
+ },
+ {
+ "epoch": 0.6549707602339181,
+ "grad_norm": 0.14377599954605103,
+ "learning_rate": 4.803918079826641e-05,
+ "loss": 0.1944,
+ "step": 560
+ },
+ {
+ "epoch": 0.6666666666666666,
+ "grad_norm": 0.44051724672317505,
+ "learning_rate": 4.7966895121363744e-05,
+ "loss": 0.1819,
+ "step": 570
+ },
+ {
+ "epoch": 0.6783625730994152,
+ "grad_norm": 0.9713970422744751,
+ "learning_rate": 4.789335745425698e-05,
+ "loss": 0.177,
+ "step": 580
+ },
+ {
+ "epoch": 0.6900584795321637,
+ "grad_norm": 0.6856667995452881,
+ "learning_rate": 4.781857180569171e-05,
+ "loss": 0.1756,
+ "step": 590
+ },
+ {
+ "epoch": 0.7017543859649122,
+ "grad_norm": 0.9721100330352783,
+ "learning_rate": 4.774254225244447e-05,
+ "loss": 0.1804,
+ "step": 600
+ },
+ {
+ "epoch": 0.7017543859649122,
+ "eval_loss": 0.17837749421596527,
+ "eval_runtime": 70.7185,
+ "eval_samples_per_second": 5.091,
+ "eval_steps_per_second": 0.636,
+ "step": 600
+ },
+ {
+ "epoch": 0.7134502923976608,
+ "grad_norm": 0.9976054430007935,
+ "learning_rate": 4.766527293910056e-05,
+ "loss": 0.182,
+ "step": 610
+ },
+ {
+ "epoch": 0.7251461988304093,
+ "grad_norm": 0.209353506565094,
+ "learning_rate": 4.75867680778281e-05,
+ "loss": 0.1955,
+ "step": 620
+ },
+ {
+ "epoch": 0.7368421052631579,
+ "grad_norm": 0.6750498414039612,
+ "learning_rate": 4.750703194814841e-05,
+ "loss": 0.1685,
+ "step": 630
+ },
+ {
+ "epoch": 0.7485380116959064,
+ "grad_norm": 0.8396773934364319,
+ "learning_rate": 4.7426068896702704e-05,
+ "loss": 0.1852,
+ "step": 640
+ },
+ {
+ "epoch": 0.7602339181286549,
+ "grad_norm": 0.5944289565086365,
+ "learning_rate": 4.734388333701519e-05,
+ "loss": 0.1782,
+ "step": 650
+ },
+ {
+ "epoch": 0.7719298245614035,
+ "grad_norm": 1.576680064201355,
+ "learning_rate": 4.726047974925244e-05,
+ "loss": 0.1661,
+ "step": 660
+ },
+ {
+ "epoch": 0.783625730994152,
+ "grad_norm": 0.5593374371528625,
+ "learning_rate": 4.717586267997916e-05,
+ "loss": 0.1638,
+ "step": 670
+ },
+ {
+ "epoch": 0.7953216374269005,
+ "grad_norm": 0.38532885909080505,
+ "learning_rate": 4.709003674191035e-05,
+ "loss": 0.1733,
+ "step": 680
+ },
+ {
+ "epoch": 0.8070175438596491,
+ "grad_norm": 0.8050347566604614,
+ "learning_rate": 4.7003006613659866e-05,
+ "loss": 0.1798,
+ "step": 690
+ },
+ {
+ "epoch": 0.8187134502923976,
+ "grad_norm": 1.1473042964935303,
+ "learning_rate": 4.691477703948537e-05,
+ "loss": 0.1647,
+ "step": 700
+ },
+ {
+ "epoch": 0.8187134502923976,
+ "eval_loss": 0.1701304167509079,
+ "eval_runtime": 70.7628,
+ "eval_samples_per_second": 5.087,
+ "eval_steps_per_second": 0.636,
+ "step": 700
+ },
+ {
+ "epoch": 0.8304093567251462,
+ "grad_norm": 0.6047437191009521,
+ "learning_rate": 4.6825352829029705e-05,
+ "loss": 0.1569,
+ "step": 710
+ },
+ {
+ "epoch": 0.8421052631578947,
+ "grad_norm": 0.7595338225364685,
+ "learning_rate": 4.673473885705869e-05,
+ "loss": 0.1748,
+ "step": 720
+ },
+ {
+ "epoch": 0.8538011695906432,
+ "grad_norm": 1.2969980239868164,
+ "learning_rate": 4.66429400631954e-05,
+ "loss": 0.2056,
+ "step": 730
+ },
+ {
+ "epoch": 0.8654970760233918,
+ "grad_norm": 0.30272266268730164,
+ "learning_rate": 4.654996145165092e-05,
+ "loss": 0.1772,
+ "step": 740
+ },
+ {
+ "epoch": 0.8771929824561403,
+ "grad_norm": 0.6495968103408813,
+ "learning_rate": 4.645580809095147e-05,
+ "loss": 0.1655,
+ "step": 750
+ },
+ {
+ "epoch": 0.8888888888888888,
+ "grad_norm": 1.118062973022461,
+ "learning_rate": 4.6360485113662216e-05,
+ "loss": 0.1826,
+ "step": 760
+ },
+ {
+ "epoch": 0.9005847953216374,
+ "grad_norm": 0.432836651802063,
+ "learning_rate": 4.626399771610739e-05,
+ "loss": 0.155,
+ "step": 770
+ },
+ {
+ "epoch": 0.9122807017543859,
+ "grad_norm": 1.9067554473876953,
+ "learning_rate": 4.616635115808705e-05,
+ "loss": 0.1791,
+ "step": 780
+ },
+ {
+ "epoch": 0.9239766081871345,
+ "grad_norm": 0.18835797905921936,
+ "learning_rate": 4.606755076259039e-05,
+ "loss": 0.167,
+ "step": 790
+ },
+ {
+ "epoch": 0.935672514619883,
+ "grad_norm": 0.27169495820999146,
+ "learning_rate": 4.5967601915505505e-05,
+ "loss": 0.1663,
+ "step": 800
+ },
+ {
+ "epoch": 0.935672514619883,
+ "eval_loss": 0.1684267818927765,
+ "eval_runtime": 70.7679,
+ "eval_samples_per_second": 5.087,
+ "eval_steps_per_second": 0.636,
+ "step": 800
+ },
+ {
+ "epoch": 0.9473684210526315,
+ "grad_norm": 0.5899501442909241,
+ "learning_rate": 4.5866510065325837e-05,
+ "loss": 0.1783,
+ "step": 810
+ },
+ {
+ "epoch": 0.9590643274853801,
+ "grad_norm": 2.2824106216430664,
+ "learning_rate": 4.576428072285316e-05,
+ "loss": 0.1765,
+ "step": 820
+ },
+ {
+ "epoch": 0.9707602339181286,
+ "grad_norm": 1.454735517501831,
+ "learning_rate": 4.5660919460897136e-05,
+ "loss": 0.1485,
+ "step": 830
+ },
+ {
+ "epoch": 0.9824561403508771,
+ "grad_norm": 0.3752526342868805,
+ "learning_rate": 4.5556431913971564e-05,
+ "loss": 0.1719,
+ "step": 840
+ },
+ {
+ "epoch": 0.9941520467836257,
+ "grad_norm": 0.4948273301124573,
+ "learning_rate": 4.5450823777987224e-05,
+ "loss": 0.1671,
+ "step": 850
+ },
+ {
+ "epoch": 1.0058479532163742,
+ "grad_norm": 0.39168354868888855,
+ "learning_rate": 4.534410080994134e-05,
+ "loss": 0.1546,
+ "step": 860
+ },
+ {
+ "epoch": 1.0175438596491229,
+ "grad_norm": 0.664540708065033,
+ "learning_rate": 4.523626882760379e-05,
+ "loss": 0.1423,
+ "step": 870
+ },
+ {
+ "epoch": 1.0292397660818713,
+ "grad_norm": 0.6726405024528503,
+ "learning_rate": 4.512733370919994e-05,
+ "loss": 0.1596,
+ "step": 880
+ },
+ {
+ "epoch": 1.04093567251462,
+ "grad_norm": 0.236578568816185,
+ "learning_rate": 4.5017301393090226e-05,
+ "loss": 0.1554,
+ "step": 890
+ },
+ {
+ "epoch": 1.0526315789473684,
+ "grad_norm": 0.3968108892440796,
+ "learning_rate": 4.490617787744641e-05,
+ "loss": 0.1553,
+ "step": 900
+ },
+ {
+ "epoch": 1.0526315789473684,
+ "eval_loss": 0.15236565470695496,
+ "eval_runtime": 70.7203,
+ "eval_samples_per_second": 5.09,
+ "eval_steps_per_second": 0.636,
+ "step": 900
+ },
+ {
+ "epoch": 1.064327485380117,
+ "grad_norm": 0.25147774815559387,
+ "learning_rate": 4.479396921992462e-05,
+ "loss": 0.1656,
+ "step": 910
+ },
+ {
+ "epoch": 1.0760233918128654,
+ "grad_norm": 0.32600972056388855,
+ "learning_rate": 4.4680681537335125e-05,
+ "loss": 0.1604,
+ "step": 920
+ },
+ {
+ "epoch": 1.087719298245614,
+ "grad_norm": 1.188389778137207,
+ "learning_rate": 4.456632100530891e-05,
+ "loss": 0.1695,
+ "step": 930
+ },
+ {
+ "epoch": 1.0994152046783625,
+ "grad_norm": 0.3761911392211914,
+ "learning_rate": 4.445089385796099e-05,
+ "loss": 0.1528,
+ "step": 940
+ },
+ {
+ "epoch": 1.1111111111111112,
+ "grad_norm": 0.3615463972091675,
+ "learning_rate": 4.4334406387550584e-05,
+ "loss": 0.1409,
+ "step": 950
+ },
+ {
+ "epoch": 1.1228070175438596,
+ "grad_norm": 1.164436936378479,
+ "learning_rate": 4.4216864944138134e-05,
+ "loss": 0.151,
+ "step": 960
+ },
+ {
+ "epoch": 1.1345029239766082,
+ "grad_norm": 0.6423844695091248,
+ "learning_rate": 4.4098275935239105e-05,
+ "loss": 0.1443,
+ "step": 970
+ },
+ {
+ "epoch": 1.1461988304093567,
+ "grad_norm": 0.5043056011199951,
+ "learning_rate": 4.39786458254747e-05,
+ "loss": 0.1723,
+ "step": 980
+ },
+ {
+ "epoch": 1.1578947368421053,
+ "grad_norm": 0.6203618049621582,
+ "learning_rate": 4.385798113621949e-05,
+ "loss": 0.1642,
+ "step": 990
+ },
+ {
+ "epoch": 1.1695906432748537,
+ "grad_norm": 1.9213874340057373,
+ "learning_rate": 4.373628844524587e-05,
+ "loss": 0.1665,
+ "step": 1000
+ },
+ {
+ "epoch": 1.1695906432748537,
+ "eval_loss": 0.15835003554821014,
+ "eval_runtime": 70.7527,
+ "eval_samples_per_second": 5.088,
+ "eval_steps_per_second": 0.636,
+ "step": 1000
+ },
+ {
+ "epoch": 1.1812865497076024,
+ "grad_norm": 0.5596243143081665,
+ "learning_rate": 4.361357438636552e-05,
+ "loss": 0.149,
+ "step": 1010
+ },
+ {
+ "epoch": 1.1929824561403508,
+ "grad_norm": 1.6553308963775635,
+ "learning_rate": 4.3489845649067756e-05,
+ "loss": 0.1487,
+ "step": 1020
+ },
+ {
+ "epoch": 1.2046783625730995,
+ "grad_norm": 1.0848426818847656,
+ "learning_rate": 4.336510897815489e-05,
+ "loss": 0.1722,
+ "step": 1030
+ },
+ {
+ "epoch": 1.2163742690058479,
+ "grad_norm": 0.7038739919662476,
+ "learning_rate": 4.323937117337451e-05,
+ "loss": 0.1586,
+ "step": 1040
+ },
+ {
+ "epoch": 1.2280701754385965,
+ "grad_norm": 0.2612368166446686,
+ "learning_rate": 4.3112639089048856e-05,
+ "loss": 0.1372,
+ "step": 1050
+ },
+ {
+ "epoch": 1.239766081871345,
+ "grad_norm": 1.2697585821151733,
+ "learning_rate": 4.298491963370115e-05,
+ "loss": 0.1375,
+ "step": 1060
+ },
+ {
+ "epoch": 1.2514619883040936,
+ "grad_norm": 0.5513612031936646,
+ "learning_rate": 4.285621976967898e-05,
+ "loss": 0.1429,
+ "step": 1070
+ },
+ {
+ "epoch": 1.263157894736842,
+ "grad_norm": 1.0210371017456055,
+ "learning_rate": 4.272654651277478e-05,
+ "loss": 0.1464,
+ "step": 1080
+ },
+ {
+ "epoch": 1.2748538011695907,
+ "grad_norm": 2.7657597064971924,
+ "learning_rate": 4.259590693184337e-05,
+ "loss": 0.143,
+ "step": 1090
+ },
+ {
+ "epoch": 1.286549707602339,
+ "grad_norm": 0.8804309964179993,
+ "learning_rate": 4.2464308148416626e-05,
+ "loss": 0.1424,
+ "step": 1100
+ },
+ {
+ "epoch": 1.286549707602339,
+ "eval_loss": 0.16600148379802704,
+ "eval_runtime": 70.6234,
+ "eval_samples_per_second": 5.097,
+ "eval_steps_per_second": 0.637,
+ "step": 1100
+ },
+ {
+ "epoch": 1.2982456140350878,
+ "grad_norm": 1.1973798274993896,
+ "learning_rate": 4.233175733631525e-05,
+ "loss": 0.147,
+ "step": 1110
+ },
+ {
+ "epoch": 1.3099415204678362,
+ "grad_norm": 0.6619781851768494,
+ "learning_rate": 4.2198261721257715e-05,
+ "loss": 0.1373,
+ "step": 1120
+ },
+ {
+ "epoch": 1.3216374269005848,
+ "grad_norm": 1.0239554643630981,
+ "learning_rate": 4.206382858046636e-05,
+ "loss": 0.1439,
+ "step": 1130
+ },
+ {
+ "epoch": 1.3333333333333333,
+ "grad_norm": 0.7170116901397705,
+ "learning_rate": 4.1928465242270696e-05,
+ "loss": 0.1372,
+ "step": 1140
+ },
+ {
+ "epoch": 1.345029239766082,
+ "grad_norm": 0.5151202082633972,
+ "learning_rate": 4.1792179085707935e-05,
+ "loss": 0.1561,
+ "step": 1150
+ },
+ {
+ "epoch": 1.3567251461988303,
+ "grad_norm": 0.5788434147834778,
+ "learning_rate": 4.1654977540120686e-05,
+ "loss": 0.1354,
+ "step": 1160
+ },
+ {
+ "epoch": 1.368421052631579,
+ "grad_norm": 3.5275087356567383,
+ "learning_rate": 4.151686808475204e-05,
+ "loss": 0.1372,
+ "step": 1170
+ },
+ {
+ "epoch": 1.3801169590643274,
+ "grad_norm": 1.1060805320739746,
+ "learning_rate": 4.137785824833776e-05,
+ "loss": 0.1776,
+ "step": 1180
+ },
+ {
+ "epoch": 1.391812865497076,
+ "grad_norm": 0.46970221400260925,
+ "learning_rate": 4.123795560869595e-05,
+ "loss": 0.1475,
+ "step": 1190
+ },
+ {
+ "epoch": 1.4035087719298245,
+ "grad_norm": 1.331180214881897,
+ "learning_rate": 4.109716779231393e-05,
+ "loss": 0.1163,
+ "step": 1200
+ },
+ {
+ "epoch": 1.4035087719298245,
+ "eval_loss": 0.14416372776031494,
+ "eval_runtime": 70.7419,
+ "eval_samples_per_second": 5.089,
+ "eval_steps_per_second": 0.636,
+ "step": 1200
+ },
+ {
+ "epoch": 1.4152046783625731,
+ "grad_norm": 0.3912939131259918,
+ "learning_rate": 4.0955502473932495e-05,
+ "loss": 0.1398,
+ "step": 1210
+ },
+ {
+ "epoch": 1.4269005847953216,
+ "grad_norm": 1.8457322120666504,
+ "learning_rate": 4.0812967376127555e-05,
+ "loss": 0.1424,
+ "step": 1220
+ },
+ {
+ "epoch": 1.4385964912280702,
+ "grad_norm": 1.9205797910690308,
+ "learning_rate": 4.0669570268889124e-05,
+ "loss": 0.125,
+ "step": 1230
+ },
+ {
+ "epoch": 1.4502923976608186,
+ "grad_norm": 0.6593519449234009,
+ "learning_rate": 4.0525318969197825e-05,
+ "loss": 0.1474,
+ "step": 1240
+ },
+ {
+ "epoch": 1.4619883040935673,
+ "grad_norm": 0.9120498895645142,
+ "learning_rate": 4.038022134059868e-05,
+ "loss": 0.1405,
+ "step": 1250
+ },
+ {
+ "epoch": 1.4736842105263157,
+ "grad_norm": 0.6781203746795654,
+ "learning_rate": 4.02342852927725e-05,
+ "loss": 0.1226,
+ "step": 1260
+ },
+ {
+ "epoch": 1.4853801169590644,
+ "grad_norm": 0.927740216255188,
+ "learning_rate": 4.00875187811047e-05,
+ "loss": 0.1362,
+ "step": 1270
+ },
+ {
+ "epoch": 1.4970760233918128,
+ "grad_norm": 0.7272914052009583,
+ "learning_rate": 3.993992980625161e-05,
+ "loss": 0.1541,
+ "step": 1280
+ },
+ {
+ "epoch": 1.5087719298245614,
+ "grad_norm": 0.3226785957813263,
+ "learning_rate": 3.979152641370434e-05,
+ "loss": 0.1429,
+ "step": 1290
+ },
+ {
+ "epoch": 1.52046783625731,
+ "grad_norm": 0.9942995309829712,
+ "learning_rate": 3.964231669335023e-05,
+ "loss": 0.1593,
+ "step": 1300
+ },
+ {
+ "epoch": 1.52046783625731,
+ "eval_loss": 0.136515274643898,
+ "eval_runtime": 70.6872,
+ "eval_samples_per_second": 5.093,
+ "eval_steps_per_second": 0.637,
+ "step": 1300
+ },
+ {
+ "epoch": 1.5321637426900585,
+ "grad_norm": 1.0689057111740112,
+ "learning_rate": 3.94923087790318e-05,
+ "loss": 0.1437,
+ "step": 1310
+ },
+ {
+ "epoch": 1.543859649122807,
+ "grad_norm": 0.9869856834411621,
+ "learning_rate": 3.934151084810338e-05,
+ "loss": 0.135,
+ "step": 1320
+ },
+ {
+ "epoch": 1.5555555555555556,
+ "grad_norm": 0.7283734679222107,
+ "learning_rate": 3.918993112098532e-05,
+ "loss": 0.1278,
+ "step": 1330
+ },
+ {
+ "epoch": 1.5672514619883042,
+ "grad_norm": 0.5857880115509033,
+ "learning_rate": 3.9037577860715915e-05,
+ "loss": 0.1384,
+ "step": 1340
+ },
+ {
+ "epoch": 1.5789473684210527,
+ "grad_norm": 1.0862607955932617,
+ "learning_rate": 3.888445937250091e-05,
+ "loss": 0.1262,
+ "step": 1350
+ },
+ {
+ "epoch": 1.590643274853801,
+ "grad_norm": 0.7065826058387756,
+ "learning_rate": 3.873058400326078e-05,
+ "loss": 0.1419,
+ "step": 1360
+ },
+ {
+ "epoch": 1.6023391812865497,
+ "grad_norm": 2.3313534259796143,
+ "learning_rate": 3.8575960141175735e-05,
+ "loss": 0.1429,
+ "step": 1370
+ },
+ {
+ "epoch": 1.6140350877192984,
+ "grad_norm": 1.8919905424118042,
+ "learning_rate": 3.842059621522843e-05,
+ "loss": 0.1402,
+ "step": 1380
+ },
+ {
+ "epoch": 1.6257309941520468,
+ "grad_norm": 0.48003846406936646,
+ "learning_rate": 3.826450069474447e-05,
+ "loss": 0.1295,
+ "step": 1390
+ },
+ {
+ "epoch": 1.6374269005847952,
+ "grad_norm": 1.175875186920166,
+ "learning_rate": 3.8107682088930794e-05,
+ "loss": 0.1529,
+ "step": 1400
+ },
+ {
+ "epoch": 1.6374269005847952,
+ "eval_loss": 0.14436422288417816,
+ "eval_runtime": 70.6551,
+ "eval_samples_per_second": 5.095,
+ "eval_steps_per_second": 0.637,
+ "step": 1400
+ },
+ {
+ "epoch": 1.6491228070175439,
+ "grad_norm": 0.714869499206543,
+ "learning_rate": 3.7950148946411703e-05,
+ "loss": 0.1212,
+ "step": 1410
+ },
+ {
+ "epoch": 1.6608187134502925,
+ "grad_norm": 0.5298812985420227,
+ "learning_rate": 3.779190985476294e-05,
+ "loss": 0.1284,
+ "step": 1420
+ },
+ {
+ "epoch": 1.672514619883041,
+ "grad_norm": 0.49085694551467896,
+ "learning_rate": 3.763297344004352e-05,
+ "loss": 0.1386,
+ "step": 1430
+ },
+ {
+ "epoch": 1.6842105263157894,
+ "grad_norm": 0.5542166829109192,
+ "learning_rate": 3.747334836632549e-05,
+ "loss": 0.1426,
+ "step": 1440
+ },
+ {
+ "epoch": 1.695906432748538,
+ "grad_norm": 1.5810950994491577,
+ "learning_rate": 3.7313043335221655e-05,
+ "loss": 0.1362,
+ "step": 1450
+ },
+ {
+ "epoch": 1.7076023391812867,
+ "grad_norm": 0.910664975643158,
+ "learning_rate": 3.715206708541122e-05,
+ "loss": 0.1498,
+ "step": 1460
+ },
+ {
+ "epoch": 1.719298245614035,
+ "grad_norm": 0.7230697870254517,
+ "learning_rate": 3.69904283921634e-05,
+ "loss": 0.1371,
+ "step": 1470
+ },
+ {
+ "epoch": 1.7309941520467835,
+ "grad_norm": 0.5364117622375488,
+ "learning_rate": 3.682813606685908e-05,
+ "loss": 0.1343,
+ "step": 1480
+ },
+ {
+ "epoch": 1.7426900584795322,
+ "grad_norm": 1.2777420282363892,
+ "learning_rate": 3.666519895651046e-05,
+ "loss": 0.1344,
+ "step": 1490
+ },
+ {
+ "epoch": 1.7543859649122808,
+ "grad_norm": 1.1144359111785889,
+ "learning_rate": 3.6501625943278805e-05,
+ "loss": 0.1321,
+ "step": 1500
+ },
+ {
+ "epoch": 1.7543859649122808,
+ "eval_loss": 0.13336236774921417,
+ "eval_runtime": 70.7727,
+ "eval_samples_per_second": 5.087,
+ "eval_steps_per_second": 0.636,
+ "step": 1500
+ },
+ {
+ "epoch": 1.7660818713450293,
+ "grad_norm": 1.9565575122833252,
+ "learning_rate": 3.6337425943990224e-05,
+ "loss": 0.1102,
+ "step": 1510
+ },
+ {
+ "epoch": 1.7777777777777777,
+ "grad_norm": 0.40069296956062317,
+ "learning_rate": 3.6172607909649606e-05,
+ "loss": 0.1352,
+ "step": 1520
+ },
+ {
+ "epoch": 1.7894736842105263,
+ "grad_norm": 0.9928420186042786,
+ "learning_rate": 3.600718082495269e-05,
+ "loss": 0.1296,
+ "step": 1530
+ },
+ {
+ "epoch": 1.801169590643275,
+ "grad_norm": 0.3943164050579071,
+ "learning_rate": 3.5841153707796245e-05,
+ "loss": 0.1427,
+ "step": 1540
+ },
+ {
+ "epoch": 1.8128654970760234,
+ "grad_norm": 0.4923238754272461,
+ "learning_rate": 3.567453560878652e-05,
+ "loss": 0.1426,
+ "step": 1550
+ },
+ {
+ "epoch": 1.8245614035087718,
+ "grad_norm": 0.944096028804779,
+ "learning_rate": 3.550733561074582e-05,
+ "loss": 0.1342,
+ "step": 1560
+ },
+ {
+ "epoch": 1.8362573099415205,
+ "grad_norm": 0.7152662873268127,
+ "learning_rate": 3.533956282821746e-05,
+ "loss": 0.1337,
+ "step": 1570
+ },
+ {
+ "epoch": 1.8479532163742691,
+ "grad_norm": 0.7894164323806763,
+ "learning_rate": 3.51712264069688e-05,
+ "loss": 0.1306,
+ "step": 1580
+ },
+ {
+ "epoch": 1.8596491228070176,
+ "grad_norm": 0.4547712206840515,
+ "learning_rate": 3.500233552349276e-05,
+ "loss": 0.1299,
+ "step": 1590
+ },
+ {
+ "epoch": 1.871345029239766,
+ "grad_norm": 2.5272104740142822,
+ "learning_rate": 3.483289938450755e-05,
+ "loss": 0.135,
+ "step": 1600
+ },
+ {
+ "epoch": 1.871345029239766,
+ "eval_loss": 0.12928339838981628,
+ "eval_runtime": 70.7709,
+ "eval_samples_per_second": 5.087,
+ "eval_steps_per_second": 0.636,
+ "step": 1600
+ },
+ {
+ "epoch": 1.8830409356725146,
+ "grad_norm": 0.702826738357544,
+ "learning_rate": 3.466292722645481e-05,
+ "loss": 0.1304,
+ "step": 1610
+ },
+ {
+ "epoch": 1.8947368421052633,
+ "grad_norm": 2.502329111099243,
+ "learning_rate": 3.449242831499608e-05,
+ "loss": 0.1203,
+ "step": 1620
+ },
+ {
+ "epoch": 1.9064327485380117,
+ "grad_norm": 0.8610795140266418,
+ "learning_rate": 3.432141194450772e-05,
+ "loss": 0.1323,
+ "step": 1630
+ },
+ {
+ "epoch": 1.9181286549707601,
+ "grad_norm": 1.4722447395324707,
+ "learning_rate": 3.4149887437574226e-05,
+ "loss": 0.1159,
+ "step": 1640
+ },
+ {
+ "epoch": 1.9298245614035088,
+ "grad_norm": 1.3295050859451294,
+ "learning_rate": 3.397786414448004e-05,
+ "loss": 0.1272,
+ "step": 1650
+ },
+ {
+ "epoch": 1.9415204678362574,
+ "grad_norm": 0.8509026765823364,
+ "learning_rate": 3.3805351442699885e-05,
+ "loss": 0.124,
+ "step": 1660
+ },
+ {
+ "epoch": 1.9532163742690059,
+ "grad_norm": 0.3647763729095459,
+ "learning_rate": 3.363235873638749e-05,
+ "loss": 0.1259,
+ "step": 1670
+ },
+ {
+ "epoch": 1.9649122807017543,
+ "grad_norm": 1.6972521543502808,
+ "learning_rate": 3.345889545586301e-05,
+ "loss": 0.1223,
+ "step": 1680
+ },
+ {
+ "epoch": 1.976608187134503,
+ "grad_norm": 0.6397933959960938,
+ "learning_rate": 3.328497105709888e-05,
+ "loss": 0.1312,
+ "step": 1690
+ },
+ {
+ "epoch": 1.9883040935672516,
+ "grad_norm": 1.981067419052124,
+ "learning_rate": 3.311059502120446e-05,
+ "loss": 0.1322,
+ "step": 1700
+ },
+ {
+ "epoch": 1.9883040935672516,
+ "eval_loss": 0.14009280502796173,
+ "eval_runtime": 70.745,
+ "eval_samples_per_second": 5.089,
+ "eval_steps_per_second": 0.636,
+ "step": 1700
+ },
+ {
+ "epoch": 2.0,
+ "grad_norm": 1.1901209354400635,
+ "learning_rate": 3.293577685390907e-05,
+ "loss": 0.1407,
+ "step": 1710
+ },
+ {
+ "epoch": 2.0116959064327484,
+ "grad_norm": 3.512953758239746,
+ "learning_rate": 3.2760526085043894e-05,
+ "loss": 0.1053,
+ "step": 1720
+ },
+ {
+ "epoch": 2.023391812865497,
+ "grad_norm": 0.8400770425796509,
+ "learning_rate": 3.258485226802242e-05,
+ "loss": 0.1356,
+ "step": 1730
+ },
+ {
+ "epoch": 2.0350877192982457,
+ "grad_norm": 0.9668805599212646,
+ "learning_rate": 3.2408764979319695e-05,
+ "loss": 0.1009,
+ "step": 1740
+ },
+ {
+ "epoch": 2.046783625730994,
+ "grad_norm": 0.7382619380950928,
+ "learning_rate": 3.2232273817950273e-05,
+ "loss": 0.1134,
+ "step": 1750
+ },
+ {
+ "epoch": 2.0584795321637426,
+ "grad_norm": 0.5820582509040833,
+ "learning_rate": 3.205538840494494e-05,
+ "loss": 0.1108,
+ "step": 1760
+ },
+ {
+ "epoch": 2.0701754385964914,
+ "grad_norm": 1.1828213930130005,
+ "learning_rate": 3.187811838282627e-05,
+ "loss": 0.1126,
+ "step": 1770
+ },
+ {
+ "epoch": 2.08187134502924,
+ "grad_norm": 0.654548168182373,
+ "learning_rate": 3.170047341508293e-05,
+ "loss": 0.1284,
+ "step": 1780
+ },
+ {
+ "epoch": 2.0935672514619883,
+ "grad_norm": 1.204846978187561,
+ "learning_rate": 3.1522463185642965e-05,
+ "loss": 0.1202,
+ "step": 1790
+ },
+ {
+ "epoch": 2.1052631578947367,
+ "grad_norm": 1.97409987449646,
+ "learning_rate": 3.134409739834584e-05,
+ "loss": 0.1355,
+ "step": 1800
+ },
+ {
+ "epoch": 2.1052631578947367,
+ "eval_loss": 0.12916673719882965,
+ "eval_runtime": 70.7041,
+ "eval_samples_per_second": 5.092,
+ "eval_steps_per_second": 0.636,
+ "step": 1800
+ },
+ {
+ "epoch": 2.116959064327485,
+ "grad_norm": 2.286799907684326,
+ "learning_rate": 3.11653857764135e-05,
+ "loss": 0.1225,
+ "step": 1810
+ },
+ {
+ "epoch": 2.128654970760234,
+ "grad_norm": 0.6355563402175903,
+ "learning_rate": 3.098633806192029e-05,
+ "loss": 0.1103,
+ "step": 1820
+ },
+ {
+ "epoch": 2.1403508771929824,
+ "grad_norm": 0.6370256543159485,
+ "learning_rate": 3.080696401526193e-05,
+ "loss": 0.1042,
+ "step": 1830
+ },
+ {
+ "epoch": 2.152046783625731,
+ "grad_norm": 1.668946385383606,
+ "learning_rate": 3.062727341462342e-05,
+ "loss": 0.1336,
+ "step": 1840
+ },
+ {
+ "epoch": 2.1637426900584797,
+ "grad_norm": 4.8773722648620605,
+ "learning_rate": 3.0447276055445984e-05,
+ "loss": 0.1212,
+ "step": 1850
+ },
+ {
+ "epoch": 2.175438596491228,
+ "grad_norm": 0.7208726406097412,
+ "learning_rate": 3.0266981749893157e-05,
+ "loss": 0.1079,
+ "step": 1860
+ },
+ {
+ "epoch": 2.1871345029239766,
+ "grad_norm": 11.563030242919922,
+ "learning_rate": 3.008640032631585e-05,
+ "loss": 0.1505,
+ "step": 1870
+ },
+ {
+ "epoch": 2.198830409356725,
+ "grad_norm": 0.6635209918022156,
+ "learning_rate": 2.990554162871656e-05,
+ "loss": 0.1306,
+ "step": 1880
+ },
+ {
+ "epoch": 2.2105263157894735,
+ "grad_norm": 0.8239225149154663,
+ "learning_rate": 2.97244155162128e-05,
+ "loss": 0.1184,
+ "step": 1890
+ },
+ {
+ "epoch": 2.2222222222222223,
+ "grad_norm": 1.2284820079803467,
+ "learning_rate": 2.954303186249962e-05,
+ "loss": 0.1227,
+ "step": 1900
+ },
+ {
+ "epoch": 2.2222222222222223,
+ "eval_loss": 0.13539400696754456,
+ "eval_runtime": 70.7773,
+ "eval_samples_per_second": 5.086,
+ "eval_steps_per_second": 0.636,
+ "step": 1900
+ },
+ {
+ "epoch": 2.2339181286549707,
+ "grad_norm": 0.5749723315238953,
+ "learning_rate": 2.936140055531135e-05,
+ "loss": 0.1208,
+ "step": 1910
+ },
+ {
+ "epoch": 2.245614035087719,
+ "grad_norm": 0.5214797854423523,
+ "learning_rate": 2.9179531495882644e-05,
+ "loss": 0.1231,
+ "step": 1920
+ },
+ {
+ "epoch": 2.257309941520468,
+ "grad_norm": 0.6751307249069214,
+ "learning_rate": 2.8997434598408653e-05,
+ "loss": 0.123,
+ "step": 1930
+ },
+ {
+ "epoch": 2.2690058479532165,
+ "grad_norm": 1.2597579956054688,
+ "learning_rate": 2.8815119789504658e-05,
+ "loss": 0.1132,
+ "step": 1940
+ },
+ {
+ "epoch": 2.280701754385965,
+ "grad_norm": 1.2100732326507568,
+ "learning_rate": 2.8632597007664895e-05,
+ "loss": 0.1049,
+ "step": 1950
+ },
+ {
+ "epoch": 2.2923976608187133,
+ "grad_norm": 0.6096538305282593,
+ "learning_rate": 2.8449876202720793e-05,
+ "loss": 0.111,
+ "step": 1960
+ },
+ {
+ "epoch": 2.3040935672514617,
+ "grad_norm": 1.1623386144638062,
+ "learning_rate": 2.8266967335298584e-05,
+ "loss": 0.1012,
+ "step": 1970
+ },
+ {
+ "epoch": 2.3157894736842106,
+ "grad_norm": 1.5342446565628052,
+ "learning_rate": 2.8083880376276307e-05,
+ "loss": 0.1121,
+ "step": 1980
+ },
+ {
+ "epoch": 2.327485380116959,
+ "grad_norm": 0.9902574419975281,
+ "learning_rate": 2.7900625306240263e-05,
+ "loss": 0.1222,
+ "step": 1990
+ },
+ {
+ "epoch": 2.3391812865497075,
+ "grad_norm": 0.4937221109867096,
+ "learning_rate": 2.7717212114940988e-05,
+ "loss": 0.1061,
+ "step": 2000
+ },
+ {
+ "epoch": 2.3391812865497075,
+ "eval_loss": 0.12341146171092987,
+ "eval_runtime": 70.7503,
+ "eval_samples_per_second": 5.088,
+ "eval_steps_per_second": 0.636,
+ "step": 2000
+ },
+ {
+ "epoch": 2.3508771929824563,
+ "grad_norm": 1.1475658416748047,
+ "learning_rate": 2.7533650800748645e-05,
+ "loss": 0.1144,
+ "step": 2010
+ },
+ {
+ "epoch": 2.3625730994152048,
+ "grad_norm": 2.0592291355133057,
+ "learning_rate": 2.7349951370107983e-05,
+ "loss": 0.1077,
+ "step": 2020
+ },
+ {
+ "epoch": 2.374269005847953,
+ "grad_norm": 1.3138481378555298,
+ "learning_rate": 2.7166123836992873e-05,
+ "loss": 0.1088,
+ "step": 2030
+ },
+ {
+ "epoch": 2.3859649122807016,
+ "grad_norm": 0.8791354894638062,
+ "learning_rate": 2.698217822236041e-05,
+ "loss": 0.1268,
+ "step": 2040
+ },
+ {
+ "epoch": 2.39766081871345,
+ "grad_norm": 1.7546913623809814,
+ "learning_rate": 2.6798124553604653e-05,
+ "loss": 0.1175,
+ "step": 2050
+ },
+ {
+ "epoch": 2.409356725146199,
+ "grad_norm": 3.894958972930908,
+ "learning_rate": 2.6613972864009994e-05,
+ "loss": 0.1177,
+ "step": 2060
+ },
+ {
+ "epoch": 2.4210526315789473,
+ "grad_norm": 0.649161696434021,
+ "learning_rate": 2.642973319220422e-05,
+ "loss": 0.1343,
+ "step": 2070
+ },
+ {
+ "epoch": 2.4327485380116958,
+ "grad_norm": 1.5214674472808838,
+ "learning_rate": 2.624541558161125e-05,
+ "loss": 0.1335,
+ "step": 2080
+ },
+ {
+ "epoch": 2.4444444444444446,
+ "grad_norm": 4.411988735198975,
+ "learning_rate": 2.606103007990371e-05,
+ "loss": 0.1208,
+ "step": 2090
+ },
+ {
+ "epoch": 2.456140350877193,
+ "grad_norm": 2.7257485389709473,
+ "learning_rate": 2.587658673845513e-05,
+ "loss": 0.1164,
+ "step": 2100
+ },
+ {
+ "epoch": 2.456140350877193,
+ "eval_loss": 0.12629590928554535,
+ "eval_runtime": 70.7902,
+ "eval_samples_per_second": 5.085,
+ "eval_steps_per_second": 0.636,
+ "step": 2100
+ },
+ {
+ "epoch": 2.4678362573099415,
+ "grad_norm": 1.7000402212142944,
+ "learning_rate": 2.569209561179206e-05,
+ "loss": 0.1408,
+ "step": 2110
+ },
+ {
+ "epoch": 2.47953216374269,
+ "grad_norm": 1.0611385107040405,
+ "learning_rate": 2.550756675704596e-05,
+ "loss": 0.1203,
+ "step": 2120
+ },
+ {
+ "epoch": 2.4912280701754383,
+ "grad_norm": 0.5716353058815002,
+ "learning_rate": 2.5323010233404925e-05,
+ "loss": 0.1187,
+ "step": 2130
+ },
+ {
+ "epoch": 2.502923976608187,
+ "grad_norm": 1.7797589302062988,
+ "learning_rate": 2.5138436101565405e-05,
+ "loss": 0.1105,
+ "step": 2140
+ },
+ {
+ "epoch": 2.5146198830409356,
+ "grad_norm": 1.4636850357055664,
+ "learning_rate": 2.4953854423183694e-05,
+ "loss": 0.1263,
+ "step": 2150
+ },
+ {
+ "epoch": 2.526315789473684,
+ "grad_norm": 1.3921252489089966,
+ "learning_rate": 2.4769275260327474e-05,
+ "loss": 0.1402,
+ "step": 2160
+ },
+ {
+ "epoch": 2.538011695906433,
+ "grad_norm": 3.662224292755127,
+ "learning_rate": 2.4584708674927308e-05,
+ "loss": 0.1254,
+ "step": 2170
+ },
+ {
+ "epoch": 2.5497076023391814,
+ "grad_norm": 1.0201853513717651,
+ "learning_rate": 2.44001647282281e-05,
+ "loss": 0.1069,
+ "step": 2180
+ },
+ {
+ "epoch": 2.56140350877193,
+ "grad_norm": 3.211782217025757,
+ "learning_rate": 2.4215653480240685e-05,
+ "loss": 0.1203,
+ "step": 2190
+ },
+ {
+ "epoch": 2.573099415204678,
+ "grad_norm": 0.7115254402160645,
+ "learning_rate": 2.4031184989193384e-05,
+ "loss": 0.1038,
+ "step": 2200
+ },
+ {
+ "epoch": 2.573099415204678,
+ "eval_loss": 0.12356504052877426,
+ "eval_runtime": 70.7673,
+ "eval_samples_per_second": 5.087,
+ "eval_steps_per_second": 0.636,
+ "step": 2200
+ },
+ {
+ "epoch": 2.5847953216374266,
+ "grad_norm": 0.9516447186470032,
+ "learning_rate": 2.3846769310983727e-05,
+ "loss": 0.1315,
+ "step": 2210
+ },
+ {
+ "epoch": 2.5964912280701755,
+ "grad_norm": 0.743248701095581,
+ "learning_rate": 2.3662416498630257e-05,
+ "loss": 0.1228,
+ "step": 2220
+ },
+ {
+ "epoch": 2.608187134502924,
+ "grad_norm": 0.5412260890007019,
+ "learning_rate": 2.3478136601724515e-05,
+ "loss": 0.115,
+ "step": 2230
+ },
+ {
+ "epoch": 2.6198830409356724,
+ "grad_norm": 0.5550256371498108,
+ "learning_rate": 2.329393966588323e-05,
+ "loss": 0.0981,
+ "step": 2240
+ },
+ {
+ "epoch": 2.6315789473684212,
+ "grad_norm": 1.205012321472168,
+ "learning_rate": 2.3109835732200686e-05,
+ "loss": 0.0954,
+ "step": 2250
+ },
+ {
+ "epoch": 2.6432748538011697,
+ "grad_norm": 0.8034434914588928,
+ "learning_rate": 2.292583483670136e-05,
+ "loss": 0.1046,
+ "step": 2260
+ },
+ {
+ "epoch": 2.654970760233918,
+ "grad_norm": 0.41518068313598633,
+ "learning_rate": 2.2741947009792817e-05,
+ "loss": 0.1245,
+ "step": 2270
+ },
+ {
+ "epoch": 2.6666666666666665,
+ "grad_norm": 1.7158474922180176,
+ "learning_rate": 2.2558182275718953e-05,
+ "loss": 0.1303,
+ "step": 2280
+ },
+ {
+ "epoch": 2.678362573099415,
+ "grad_norm": 2.198917865753174,
+ "learning_rate": 2.2374550652013492e-05,
+ "loss": 0.1142,
+ "step": 2290
+ },
+ {
+ "epoch": 2.690058479532164,
+ "grad_norm": 2.116436719894409,
+ "learning_rate": 2.219106214895397e-05,
+ "loss": 0.1196,
+ "step": 2300
+ },
+ {
+ "epoch": 2.690058479532164,
+ "eval_loss": 0.12174443155527115,
+ "eval_runtime": 70.7411,
+ "eval_samples_per_second": 5.089,
+ "eval_steps_per_second": 0.636,
+ "step": 2300
+ },
+ {
+ "epoch": 2.7017543859649122,
+ "grad_norm": 1.5849992036819458,
+ "learning_rate": 2.2007726769016006e-05,
+ "loss": 0.1071,
+ "step": 2310
+ },
+ {
+ "epoch": 2.7134502923976607,
+ "grad_norm": 3.807744026184082,
+ "learning_rate": 2.182455450632803e-05,
+ "loss": 0.1246,
+ "step": 2320
+ },
+ {
+ "epoch": 2.7251461988304095,
+ "grad_norm": 1.429018497467041,
+ "learning_rate": 2.1641555346126512e-05,
+ "loss": 0.1188,
+ "step": 2330
+ },
+ {
+ "epoch": 2.736842105263158,
+ "grad_norm": 0.38770321011543274,
+ "learning_rate": 2.1458739264211618e-05,
+ "loss": 0.101,
+ "step": 2340
+ },
+ {
+ "epoch": 2.7485380116959064,
+ "grad_norm": 1.6713892221450806,
+ "learning_rate": 2.127611622640336e-05,
+ "loss": 0.1068,
+ "step": 2350
+ },
+ {
+ "epoch": 2.760233918128655,
+ "grad_norm": 1.5625298023223877,
+ "learning_rate": 2.1093696187998424e-05,
+ "loss": 0.1196,
+ "step": 2360
+ },
+ {
+ "epoch": 2.7719298245614032,
+ "grad_norm": 0.6439236402511597,
+ "learning_rate": 2.0911489093227395e-05,
+ "loss": 0.1076,
+ "step": 2370
+ },
+ {
+ "epoch": 2.783625730994152,
+ "grad_norm": 0.947961151599884,
+ "learning_rate": 2.0729504874712713e-05,
+ "loss": 0.111,
+ "step": 2380
+ },
+ {
+ "epoch": 2.7953216374269005,
+ "grad_norm": 0.6834493279457092,
+ "learning_rate": 2.0547753452927197e-05,
+ "loss": 0.1079,
+ "step": 2390
+ },
+ {
+ "epoch": 2.807017543859649,
+ "grad_norm": 0.9059198498725891,
+ "learning_rate": 2.036624473565328e-05,
+ "loss": 0.1058,
+ "step": 2400
+ },
+ {
+ "epoch": 2.807017543859649,
+ "eval_loss": 0.1205640360713005,
+ "eval_runtime": 70.758,
+ "eval_samples_per_second": 5.088,
+ "eval_steps_per_second": 0.636,
+ "step": 2400
+ },
+ {
+ "epoch": 2.818713450292398,
+ "grad_norm": 1.0221079587936401,
+ "learning_rate": 2.0184988617442852e-05,
+ "loss": 0.1111,
+ "step": 2410
+ },
+ {
+ "epoch": 2.8304093567251463,
+ "grad_norm": 1.409433126449585,
+ "learning_rate": 2.0003994979077937e-05,
+ "loss": 0.1069,
+ "step": 2420
+ },
+ {
+ "epoch": 2.8421052631578947,
+ "grad_norm": 3.338195562362671,
+ "learning_rate": 1.9823273687032045e-05,
+ "loss": 0.1153,
+ "step": 2430
+ },
+ {
+ "epoch": 2.853801169590643,
+ "grad_norm": 1.3215330839157104,
+ "learning_rate": 1.964283459293232e-05,
+ "loss": 0.1097,
+ "step": 2440
+ },
+ {
+ "epoch": 2.8654970760233915,
+ "grad_norm": 0.9754563570022583,
+ "learning_rate": 1.94626875330225e-05,
+ "loss": 0.1123,
+ "step": 2450
+ },
+ {
+ "epoch": 2.8771929824561404,
+ "grad_norm": 1.7067068815231323,
+ "learning_rate": 1.92828423276267e-05,
+ "loss": 0.1194,
+ "step": 2460
+ },
+ {
+ "epoch": 2.888888888888889,
+ "grad_norm": 0.7298212051391602,
+ "learning_rate": 1.9103308780614118e-05,
+ "loss": 0.1041,
+ "step": 2470
+ },
+ {
+ "epoch": 2.9005847953216373,
+ "grad_norm": 2.4480433464050293,
+ "learning_rate": 1.8924096678864557e-05,
+ "loss": 0.1155,
+ "step": 2480
+ },
+ {
+ "epoch": 2.912280701754386,
+ "grad_norm": 0.9016093611717224,
+ "learning_rate": 1.8745215791734942e-05,
+ "loss": 0.1073,
+ "step": 2490
+ },
+ {
+ "epoch": 2.9239766081871346,
+ "grad_norm": 1.3098756074905396,
+ "learning_rate": 1.8566675870526756e-05,
+ "loss": 0.1177,
+ "step": 2500
+ },
+ {
+ "epoch": 2.9239766081871346,
+ "eval_loss": 0.12042708694934845,
+ "eval_runtime": 70.7595,
+ "eval_samples_per_second": 5.088,
+ "eval_steps_per_second": 0.636,
+ "step": 2500
+ },
+ {
+ "epoch": 2.935672514619883,
+ "grad_norm": 0.7243450284004211,
+ "learning_rate": 1.838848664795447e-05,
+ "loss": 0.1073,
+ "step": 2510
+ },
+ {
+ "epoch": 2.9473684210526314,
+ "grad_norm": 0.7760084271430969,
+ "learning_rate": 1.8210657837614965e-05,
+ "loss": 0.11,
+ "step": 2520
+ },
+ {
+ "epoch": 2.95906432748538,
+ "grad_norm": 3.328228712081909,
+ "learning_rate": 1.803319913345805e-05,
+ "loss": 0.112,
+ "step": 2530
+ },
+ {
+ "epoch": 2.9707602339181287,
+ "grad_norm": 2.186782121658325,
+ "learning_rate": 1.785612020925801e-05,
+ "loss": 0.0969,
+ "step": 2540
+ },
+ {
+ "epoch": 2.982456140350877,
+ "grad_norm": 0.893616259098053,
+ "learning_rate": 1.7679430718086243e-05,
+ "loss": 0.1006,
+ "step": 2550
+ },
+ {
+ "epoch": 2.9941520467836256,
+ "grad_norm": 3.7202115058898926,
+ "learning_rate": 1.7503140291785054e-05,
+ "loss": 0.1041,
+ "step": 2560
+ },
+ {
+ "epoch": 3.0058479532163744,
+ "grad_norm": 1.9740840196609497,
+ "learning_rate": 1.732725854044261e-05,
+ "loss": 0.0973,
+ "step": 2570
+ },
+ {
+ "epoch": 3.017543859649123,
+ "grad_norm": 3.9907517433166504,
+ "learning_rate": 1.7151795051869025e-05,
+ "loss": 0.1181,
+ "step": 2580
+ },
+ {
+ "epoch": 3.0292397660818713,
+ "grad_norm": 0.8175060153007507,
+ "learning_rate": 1.6976759391073758e-05,
+ "loss": 0.1177,
+ "step": 2590
+ },
+ {
+ "epoch": 3.0409356725146197,
+ "grad_norm": 1.1770973205566406,
+ "learning_rate": 1.6802161099744164e-05,
+ "loss": 0.1009,
+ "step": 2600
+ },
+ {
+ "epoch": 3.0409356725146197,
+ "eval_loss": 0.11801114678382874,
+ "eval_runtime": 70.7625,
+ "eval_samples_per_second": 5.087,
+ "eval_steps_per_second": 0.636,
+ "step": 2600
+ },
+ {
+ "epoch": 3.0526315789473686,
+ "grad_norm": 0.982765793800354,
+ "learning_rate": 1.6628009695725346e-05,
+ "loss": 0.0869,
+ "step": 2610
+ },
+ {
+ "epoch": 3.064327485380117,
+ "grad_norm": 1.3147976398468018,
+ "learning_rate": 1.645431467250134e-05,
+ "loss": 0.1067,
+ "step": 2620
+ },
+ {
+ "epoch": 3.0760233918128654,
+ "grad_norm": 2.452866315841675,
+ "learning_rate": 1.628108549867754e-05,
+ "loss": 0.1065,
+ "step": 2630
+ },
+ {
+ "epoch": 3.087719298245614,
+ "grad_norm": 1.246812105178833,
+ "learning_rate": 1.610833161746461e-05,
+ "loss": 0.1046,
+ "step": 2640
+ },
+ {
+ "epoch": 3.0994152046783627,
+ "grad_norm": 2.475942611694336,
+ "learning_rate": 1.5936062446163666e-05,
+ "loss": 0.102,
+ "step": 2650
+ },
+ {
+ "epoch": 3.111111111111111,
+ "grad_norm": 2.2378146648406982,
+ "learning_rate": 1.576428737565292e-05,
+ "loss": 0.0921,
+ "step": 2660
+ },
+ {
+ "epoch": 3.1228070175438596,
+ "grad_norm": 1.563463807106018,
+ "learning_rate": 1.5593015769875746e-05,
+ "loss": 0.0966,
+ "step": 2670
+ },
+ {
+ "epoch": 3.134502923976608,
+ "grad_norm": 2.7009105682373047,
+ "learning_rate": 1.5422256965330263e-05,
+ "loss": 0.1179,
+ "step": 2680
+ },
+ {
+ "epoch": 3.146198830409357,
+ "grad_norm": 0.5407319664955139,
+ "learning_rate": 1.52520202705603e-05,
+ "loss": 0.095,
+ "step": 2690
+ },
+ {
+ "epoch": 3.1578947368421053,
+ "grad_norm": 1.054494857788086,
+ "learning_rate": 1.508231496564806e-05,
+ "loss": 0.1024,
+ "step": 2700
+ },
+ {
+ "epoch": 3.1578947368421053,
+ "eval_loss": 0.11845976859331131,
+ "eval_runtime": 70.8082,
+ "eval_samples_per_second": 5.084,
+ "eval_steps_per_second": 0.636,
+ "step": 2700
+ },
+ {
+ "epoch": 3.1695906432748537,
+ "grad_norm": 1.0297489166259766,
+ "learning_rate": 1.4913150301708162e-05,
+ "loss": 0.1048,
+ "step": 2710
+ },
+ {
+ "epoch": 3.181286549707602,
+ "grad_norm": 7.1104254722595215,
+ "learning_rate": 1.4744535500383372e-05,
+ "loss": 0.1227,
+ "step": 2720
+ },
+ {
+ "epoch": 3.192982456140351,
+ "grad_norm": 1.7294998168945312,
+ "learning_rate": 1.4576479753341882e-05,
+ "loss": 0.1155,
+ "step": 2730
+ },
+ {
+ "epoch": 3.2046783625730995,
+ "grad_norm": 1.502966046333313,
+ "learning_rate": 1.4408992221776282e-05,
+ "loss": 0.1116,
+ "step": 2740
+ },
+ {
+ "epoch": 3.216374269005848,
+ "grad_norm": 1.003847360610962,
+ "learning_rate": 1.4242082035904086e-05,
+ "loss": 0.1168,
+ "step": 2750
+ },
+ {
+ "epoch": 3.2280701754385963,
+ "grad_norm": 1.5059524774551392,
+ "learning_rate": 1.4075758294470132e-05,
+ "loss": 0.0942,
+ "step": 2760
+ },
+ {
+ "epoch": 3.239766081871345,
+ "grad_norm": 0.9065340161323547,
+ "learning_rate": 1.3910030064250462e-05,
+ "loss": 0.103,
+ "step": 2770
+ },
+ {
+ "epoch": 3.2514619883040936,
+ "grad_norm": 1.1417133808135986,
+ "learning_rate": 1.3744906379558165e-05,
+ "loss": 0.1246,
+ "step": 2780
+ },
+ {
+ "epoch": 3.263157894736842,
+ "grad_norm": 1.5030591487884521,
+ "learning_rate": 1.3580396241750823e-05,
+ "loss": 0.1054,
+ "step": 2790
+ },
+ {
+ "epoch": 3.2748538011695905,
+ "grad_norm": 2.610001564025879,
+ "learning_rate": 1.3416508618739854e-05,
+ "loss": 0.0953,
+ "step": 2800
+ },
+ {
+ "epoch": 3.2748538011695905,
+ "eval_loss": 0.1236434280872345,
+ "eval_runtime": 70.7784,
+ "eval_samples_per_second": 5.086,
+ "eval_steps_per_second": 0.636,
+ "step": 2800
+ },
+ {
+ "epoch": 3.2865497076023393,
+ "grad_norm": 1.7465177774429321,
+ "learning_rate": 1.3253252444501635e-05,
+ "loss": 0.0891,
+ "step": 2810
+ },
+ {
+ "epoch": 3.2982456140350878,
+ "grad_norm": 1.5508021116256714,
+ "learning_rate": 1.309063661859049e-05,
+ "loss": 0.0942,
+ "step": 2820
+ },
+ {
+ "epoch": 3.309941520467836,
+ "grad_norm": 0.9697007536888123,
+ "learning_rate": 1.2928670005653576e-05,
+ "loss": 0.0966,
+ "step": 2830
+ },
+ {
+ "epoch": 3.3216374269005846,
+ "grad_norm": 1.3551596403121948,
+ "learning_rate": 1.2767361434947584e-05,
+ "loss": 0.0863,
+ "step": 2840
+ },
+ {
+ "epoch": 3.3333333333333335,
+ "grad_norm": 2.1799612045288086,
+ "learning_rate": 1.2606719699857489e-05,
+ "loss": 0.0938,
+ "step": 2850
+ },
+ {
+ "epoch": 3.345029239766082,
+ "grad_norm": 3.5059502124786377,
+ "learning_rate": 1.244675355741717e-05,
+ "loss": 0.1161,
+ "step": 2860
+ },
+ {
+ "epoch": 3.3567251461988303,
+ "grad_norm": 1.9047223329544067,
+ "learning_rate": 1.228747172783204e-05,
+ "loss": 0.1082,
+ "step": 2870
+ },
+ {
+ "epoch": 3.3684210526315788,
+ "grad_norm": 0.8077034950256348,
+ "learning_rate": 1.212888289400371e-05,
+ "loss": 0.0982,
+ "step": 2880
+ },
+ {
+ "epoch": 3.3801169590643276,
+ "grad_norm": 4.194718360900879,
+ "learning_rate": 1.1970995701056615e-05,
+ "loss": 0.1026,
+ "step": 2890
+ },
+ {
+ "epoch": 3.391812865497076,
+ "grad_norm": 1.0910911560058594,
+ "learning_rate": 1.181381875586677e-05,
+ "loss": 0.0905,
+ "step": 2900
+ },
+ {
+ "epoch": 3.391812865497076,
+ "eval_loss": 0.12022969871759415,
+ "eval_runtime": 70.7539,
+ "eval_samples_per_second": 5.088,
+ "eval_steps_per_second": 0.636,
+ "step": 2900
+ },
+ {
+ "epoch": 3.391812865497076,
+ "step": 2900,
+ "total_flos": 5.067618607815459e+18,
+ "train_loss": 0.14973004334959492,
+ "train_runtime": 17301.2227,
+ "train_samples_per_second": 1.977,
+ "train_steps_per_second": 0.247
+ }
+ ],
+ "logging_steps": 10,
+ "max_steps": 4275,
+ "num_input_tokens_seen": 0,
+ "num_train_epochs": 5,
+ "save_steps": 100,
+ "total_flos": 5.067618607815459e+18,
+ "train_batch_size": 4,
+ "trial_name": null,
+ "trial_params": null
+}
diff --git a/llama2_13b_peft/paragraph_segmentation/training_args.bin b/llama2_13b_peft/paragraph_segmentation/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..bdefa332c2174d194c5a67f963c83c5183601953
--- /dev/null
+++ b/llama2_13b_peft/paragraph_segmentation/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b46a67d8364305ef66b05e699b604996db4a79a12ca30354ea647391b8a5316d
+size 5176
diff --git a/llama2_13b_peft/paragraph_segmentation/training_eval_loss.png b/llama2_13b_peft/paragraph_segmentation/training_eval_loss.png
new file mode 100644
index 0000000000000000000000000000000000000000..d13be1b292939f4f55944179c9c1ca0c01e377c4
Binary files /dev/null and b/llama2_13b_peft/paragraph_segmentation/training_eval_loss.png differ
diff --git a/llama2_13b_peft/paragraph_segmentation/training_loss.png b/llama2_13b_peft/paragraph_segmentation/training_loss.png
new file mode 100644
index 0000000000000000000000000000000000000000..fb543efa0c055f985213bd03f2914f210856e9ff
Binary files /dev/null and b/llama2_13b_peft/paragraph_segmentation/training_loss.png differ
diff --git a/llama2_13b_peft/play_dialog_same_or_different/README.md b/llama2_13b_peft/play_dialog_same_or_different/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..d76f5a0478ba3a293951faa6c4f436f457bcc562
--- /dev/null
+++ b/llama2_13b_peft/play_dialog_same_or_different/README.md
@@ -0,0 +1,73 @@
+---
+license: other
+library_name: peft
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+base_model: /data1/model/llama2/meta-llama/Llama2-13b
+model-index:
+- name: play_dialog_same_or_different_no_sys
+ results: []
+---
+
+
+
+# play_dialog_same_or_different_no_sys
+
+This model is a fine-tuned version of [/data1/model/llama2/meta-llama/Llama2-13b](https://huggingface.co//data1/model/llama2/meta-llama/Llama2-13b) on the play_dialog_same_or_different_no_sys dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.3175
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 4
+- eval_batch_size: 4
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 2
+- total_train_batch_size: 8
+- total_eval_batch_size: 8
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_steps: 20
+- num_epochs: 5.0
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:------:|:----:|:---------------:|
+| 0.3588 | 0.3215 | 100 | 0.3202 |
+| 0.321 | 0.6431 | 200 | 0.3241 |
+| 0.3268 | 0.9646 | 300 | 0.3193 |
+| 0.34 | 1.2862 | 400 | 0.3250 |
+| 0.3221 | 1.6077 | 500 | 0.3175 |
+| 0.3193 | 1.9293 | 600 | 0.3194 |
+| 0.339 | 2.2508 | 700 | 0.3211 |
+| 0.3347 | 2.5723 | 800 | 0.3262 |
+
+
+### Framework versions
+
+- PEFT 0.10.0
+- Transformers 4.40.0
+- Pytorch 2.2.1
+- Datasets 2.18.0
+- Tokenizers 0.19.1
\ No newline at end of file
diff --git a/llama2_13b_peft/play_dialog_same_or_different/adapter_config.json b/llama2_13b_peft/play_dialog_same_or_different/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..80a64a9bea5e22f8b4561357dcc0838f7fa5a39a
--- /dev/null
+++ b/llama2_13b_peft/play_dialog_same_or_different/adapter_config.json
@@ -0,0 +1,34 @@
+{
+ "alpha_pattern": {},
+ "auto_mapping": null,
+ "base_model_name_or_path": "/data1/model/llama2/meta-llama/Llama2-13b",
+ "bias": "none",
+ "fan_in_fan_out": false,
+ "inference_mode": true,
+ "init_lora_weights": true,
+ "layer_replication": null,
+ "layers_pattern": null,
+ "layers_to_transform": null,
+ "loftq_config": {},
+ "lora_alpha": 16,
+ "lora_dropout": 0.0,
+ "megatron_config": null,
+ "megatron_core": "megatron.core",
+ "modules_to_save": null,
+ "peft_type": "LORA",
+ "r": 8,
+ "rank_pattern": {},
+ "revision": null,
+ "target_modules": [
+ "down_proj",
+ "k_proj",
+ "gate_proj",
+ "up_proj",
+ "o_proj",
+ "q_proj",
+ "v_proj"
+ ],
+ "task_type": "CAUSAL_LM",
+ "use_dora": false,
+ "use_rslora": false
+}
\ No newline at end of file
diff --git a/llama2_13b_peft/play_dialog_same_or_different/adapter_model.safetensors b/llama2_13b_peft/play_dialog_same_or_different/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..77a352233461f2a25804249299e09d7725c5eed8
--- /dev/null
+++ b/llama2_13b_peft/play_dialog_same_or_different/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c8c83ac7a47be01b3b27a6d3b9aae775e94b660b6e0717369fa2c9340ce3144
+size 125248064
diff --git a/llama2_13b_peft/play_dialog_same_or_different/all_results.json b/llama2_13b_peft/play_dialog_same_or_different/all_results.json
new file mode 100644
index 0000000000000000000000000000000000000000..efa7562edd4ea16be5abdde6b5b73bef66a95ace
--- /dev/null
+++ b/llama2_13b_peft/play_dialog_same_or_different/all_results.json
@@ -0,0 +1,12 @@
+{
+ "epoch": 2.572347266881029,
+ "eval_loss": 0.31748995184898376,
+ "eval_runtime": 2.9141,
+ "eval_samples_per_second": 44.953,
+ "eval_steps_per_second": 5.834,
+ "total_flos": 1.4161342881595392e+17,
+ "train_loss": 0.5724774128198624,
+ "train_runtime": 490.2786,
+ "train_samples_per_second": 25.302,
+ "train_steps_per_second": 3.172
+}
\ No newline at end of file
diff --git a/llama2_13b_peft/play_dialog_same_or_different/eval_results.json b/llama2_13b_peft/play_dialog_same_or_different/eval_results.json
new file mode 100644
index 0000000000000000000000000000000000000000..b8f566aece749ed499607646696c667506947f26
--- /dev/null
+++ b/llama2_13b_peft/play_dialog_same_or_different/eval_results.json
@@ -0,0 +1,7 @@
+{
+ "epoch": 2.572347266881029,
+ "eval_loss": 0.31748995184898376,
+ "eval_runtime": 2.9141,
+ "eval_samples_per_second": 44.953,
+ "eval_steps_per_second": 5.834
+}
\ No newline at end of file
diff --git a/llama2_13b_peft/play_dialog_same_or_different/special_tokens_map.json b/llama2_13b_peft/play_dialog_same_or_different/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..72ecfeeb7e14d244c936169d2ed139eeae235ef1
--- /dev/null
+++ b/llama2_13b_peft/play_dialog_same_or_different/special_tokens_map.json
@@ -0,0 +1,24 @@
+{
+ "bos_token": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false
+ },
+ "eos_token": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false
+ },
+ "pad_token": "",
+ "unk_token": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false
+ }
+}
diff --git a/llama2_13b_peft/play_dialog_same_or_different/tokenizer.model b/llama2_13b_peft/play_dialog_same_or_different/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..6c00c742ce03c627d6cd5b795984876fa49fa899
--- /dev/null
+++ b/llama2_13b_peft/play_dialog_same_or_different/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723
diff --git a/llama2_13b_peft/play_dialog_same_or_different/tokenizer_config.json b/llama2_13b_peft/play_dialog_same_or_different/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..a40266f39e5b5fed14de34710d35eb9e98d6bdad
--- /dev/null
+++ b/llama2_13b_peft/play_dialog_same_or_different/tokenizer_config.json
@@ -0,0 +1,45 @@
+{
+ "add_bos_token": true,
+ "add_eos_token": false,
+ "add_prefix_space": true,
+ "added_tokens_decoder": {
+ "0": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false,
+ "special": true
+ },
+ "1": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false,
+ "special": true
+ },
+ "2": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false,
+ "special": true
+ }
+ },
+ "bos_token": "",
+ "chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '\\n' }}{% endif %}{% endfor %}",
+ "clean_up_tokenization_spaces": false,
+ "eos_token": "",
+ "legacy": true,
+ "model_max_length": 1000000000000000019884624838656,
+ "pad_token": "",
+ "padding_side": "right",
+ "sp_model_kwargs": {},
+ "spaces_between_special_tokens": false,
+ "split_special_tokens": false,
+ "tokenizer_class": "LlamaTokenizer",
+ "unk_token": "",
+ "use_default_system_prompt": false
+}
diff --git a/llama2_13b_peft/play_dialog_same_or_different/train_results.json b/llama2_13b_peft/play_dialog_same_or_different/train_results.json
new file mode 100644
index 0000000000000000000000000000000000000000..b26cdfe75aedf85af2d86e5083beec3ec0823f2f
--- /dev/null
+++ b/llama2_13b_peft/play_dialog_same_or_different/train_results.json
@@ -0,0 +1,8 @@
+{
+ "epoch": 2.572347266881029,
+ "total_flos": 1.4161342881595392e+17,
+ "train_loss": 0.5724774128198624,
+ "train_runtime": 490.2786,
+ "train_samples_per_second": 25.302,
+ "train_steps_per_second": 3.172
+}
\ No newline at end of file
diff --git a/llama2_13b_peft/play_dialog_same_or_different/trainer_log.jsonl b/llama2_13b_peft/play_dialog_same_or_different/trainer_log.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..52a9949ddf949bf86dddf19bd4bf6fecc95e42dc
--- /dev/null
+++ b/llama2_13b_peft/play_dialog_same_or_different/trainer_log.jsonl
@@ -0,0 +1,90 @@
+{"current_steps": 10, "total_steps": 1555, "loss": 10.7274, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.5e-05, "epoch": 0.03215434083601286, "percentage": 0.64, "elapsed_time": "0:00:07", "remaining_time": "0:18:20"}
+{"current_steps": 20, "total_steps": 1555, "loss": 7.7527, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 5e-05, "epoch": 0.06430868167202572, "percentage": 1.29, "elapsed_time": "0:00:12", "remaining_time": "0:16:00"}
+{"current_steps": 30, "total_steps": 1555, "loss": 1.1271, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.999476426280588e-05, "epoch": 0.09646302250803858, "percentage": 1.93, "elapsed_time": "0:00:18", "remaining_time": "0:15:22"}
+{"current_steps": 40, "total_steps": 1555, "loss": 0.4741, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.997905924425903e-05, "epoch": 0.12861736334405144, "percentage": 2.57, "elapsed_time": "0:00:23", "remaining_time": "0:14:57"}
+{"current_steps": 50, "total_steps": 1555, "loss": 0.3836, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.995289152254744e-05, "epoch": 0.1607717041800643, "percentage": 3.22, "elapsed_time": "0:00:29", "remaining_time": "0:14:45"}
+{"current_steps": 60, "total_steps": 1555, "loss": 0.3745, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.991627205825621e-05, "epoch": 0.19292604501607716, "percentage": 3.86, "elapsed_time": "0:00:34", "remaining_time": "0:14:30"}
+{"current_steps": 70, "total_steps": 1555, "loss": 0.3658, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.986921618977664e-05, "epoch": 0.22508038585209003, "percentage": 4.5, "elapsed_time": "0:00:40", "remaining_time": "0:14:27"}
+{"current_steps": 80, "total_steps": 1555, "loss": 0.346, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.981174362688158e-05, "epoch": 0.2572347266881029, "percentage": 5.14, "elapsed_time": "0:00:46", "remaining_time": "0:14:19"}
+{"current_steps": 90, "total_steps": 1555, "loss": 0.3624, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.974387844246987e-05, "epoch": 0.28938906752411575, "percentage": 5.79, "elapsed_time": "0:00:52", "remaining_time": "0:14:06"}
+{"current_steps": 100, "total_steps": 1555, "loss": 0.3588, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.9665649062483115e-05, "epoch": 0.3215434083601286, "percentage": 6.43, "elapsed_time": "0:00:57", "remaining_time": "0:13:56"}
+{"current_steps": 100, "total_steps": 1555, "loss": null, "eval_loss": 0.3202289044857025, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 0.3215434083601286, "percentage": 6.43, "elapsed_time": "0:00:57", "remaining_time": "0:13:56"}
+{"current_steps": 110, "total_steps": 1555, "loss": 0.341, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.957708825399927e-05, "epoch": 0.3536977491961415, "percentage": 7.07, "elapsed_time": "0:01:06", "remaining_time": "0:14:34"}
+{"current_steps": 120, "total_steps": 1555, "loss": 0.3741, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.9478233111507856e-05, "epoch": 0.3858520900321543, "percentage": 7.72, "elapsed_time": "0:01:12", "remaining_time": "0:14:21"}
+{"current_steps": 130, "total_steps": 1555, "loss": 0.3254, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.936912504137257e-05, "epoch": 0.4180064308681672, "percentage": 8.36, "elapsed_time": "0:01:17", "remaining_time": "0:14:09"}
+{"current_steps": 140, "total_steps": 1555, "loss": 0.3496, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.924980974448791e-05, "epoch": 0.45016077170418006, "percentage": 9.0, "elapsed_time": "0:01:23", "remaining_time": "0:14:01"}
+{"current_steps": 150, "total_steps": 1555, "loss": 0.3569, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.912033719713687e-05, "epoch": 0.48231511254019294, "percentage": 9.65, "elapsed_time": "0:01:28", "remaining_time": "0:13:51"}
+{"current_steps": 160, "total_steps": 1555, "loss": 0.3066, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.8980761630058014e-05, "epoch": 0.5144694533762058, "percentage": 10.29, "elapsed_time": "0:01:34", "remaining_time": "0:13:42"}
+{"current_steps": 170, "total_steps": 1555, "loss": 0.356, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.883114150573037e-05, "epoch": 0.5466237942122186, "percentage": 10.93, "elapsed_time": "0:01:40", "remaining_time": "0:13:37"}
+{"current_steps": 180, "total_steps": 1555, "loss": 0.3955, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.8671539493885934e-05, "epoch": 0.5787781350482315, "percentage": 11.58, "elapsed_time": "0:01:46", "remaining_time": "0:13:29"}
+{"current_steps": 190, "total_steps": 1555, "loss": 0.3876, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.850202244525987e-05, "epoch": 0.6109324758842444, "percentage": 12.22, "elapsed_time": "0:01:51", "remaining_time": "0:13:22"}
+{"current_steps": 200, "total_steps": 1555, "loss": 0.321, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.8322661363589507e-05, "epoch": 0.6430868167202572, "percentage": 12.86, "elapsed_time": "0:01:57", "remaining_time": "0:13:15"}
+{"current_steps": 200, "total_steps": 1555, "loss": null, "eval_loss": 0.32409238815307617, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 0.6430868167202572, "percentage": 12.86, "elapsed_time": "0:01:57", "remaining_time": "0:13:15"}
+{"current_steps": 210, "total_steps": 1555, "loss": 0.39, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.813353137587377e-05, "epoch": 0.6752411575562701, "percentage": 13.5, "elapsed_time": "0:02:06", "remaining_time": "0:13:29"}
+{"current_steps": 220, "total_steps": 1555, "loss": 0.3429, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.793471170090555e-05, "epoch": 0.707395498392283, "percentage": 14.15, "elapsed_time": "0:02:12", "remaining_time": "0:13:22"}
+{"current_steps": 230, "total_steps": 1555, "loss": 0.3208, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.772628561609022e-05, "epoch": 0.7395498392282959, "percentage": 14.79, "elapsed_time": "0:02:17", "remaining_time": "0:13:13"}
+{"current_steps": 240, "total_steps": 1555, "loss": 0.344, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.7508340422564134e-05, "epoch": 0.7717041800643086, "percentage": 15.43, "elapsed_time": "0:02:23", "remaining_time": "0:13:05"}
+{"current_steps": 250, "total_steps": 1555, "loss": 0.2976, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.728096740862778e-05, "epoch": 0.8038585209003215, "percentage": 16.08, "elapsed_time": "0:02:29", "remaining_time": "0:12:59"}
+{"current_steps": 260, "total_steps": 1555, "loss": 0.3684, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.704426181150884e-05, "epoch": 0.8360128617363344, "percentage": 16.72, "elapsed_time": "0:02:34", "remaining_time": "0:12:51"}
+{"current_steps": 270, "total_steps": 1555, "loss": 0.3398, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.6798322777471216e-05, "epoch": 0.8681672025723473, "percentage": 17.36, "elapsed_time": "0:02:40", "remaining_time": "0:12:43"}
+{"current_steps": 280, "total_steps": 1555, "loss": 0.3236, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.654325332028676e-05, "epoch": 0.9003215434083601, "percentage": 18.01, "elapsed_time": "0:02:46", "remaining_time": "0:12:37"}
+{"current_steps": 290, "total_steps": 1555, "loss": 0.2567, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.6279160278086994e-05, "epoch": 0.932475884244373, "percentage": 18.65, "elapsed_time": "0:02:51", "remaining_time": "0:12:30"}
+{"current_steps": 300, "total_steps": 1555, "loss": 0.3268, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.6006154268613014e-05, "epoch": 0.9646302250803859, "percentage": 19.29, "elapsed_time": "0:02:57", "remaining_time": "0:12:22"}
+{"current_steps": 300, "total_steps": 1555, "loss": null, "eval_loss": 0.3192877471446991, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 0.9646302250803859, "percentage": 19.29, "elapsed_time": "0:02:57", "remaining_time": "0:12:22"}
+{"current_steps": 310, "total_steps": 1555, "loss": 0.3136, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.572434964288226e-05, "epoch": 0.9967845659163987, "percentage": 19.94, "elapsed_time": "0:03:06", "remaining_time": "0:12:28"}
+{"current_steps": 320, "total_steps": 1555, "loss": 0.3386, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.543386443729157e-05, "epoch": 1.0289389067524115, "percentage": 20.58, "elapsed_time": "0:03:11", "remaining_time": "0:12:20"}
+{"current_steps": 330, "total_steps": 1555, "loss": 0.3385, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.513482032417656e-05, "epoch": 1.0610932475884245, "percentage": 21.22, "elapsed_time": "0:03:17", "remaining_time": "0:12:14"}
+{"current_steps": 340, "total_steps": 1555, "loss": 0.3345, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.48273425608481e-05, "epoch": 1.0932475884244373, "percentage": 21.86, "elapsed_time": "0:03:23", "remaining_time": "0:12:07"}
+{"current_steps": 350, "total_steps": 1555, "loss": 0.3515, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.451155993712711e-05, "epoch": 1.1254019292604502, "percentage": 22.51, "elapsed_time": "0:03:29", "remaining_time": "0:11:59"}
+{"current_steps": 360, "total_steps": 1555, "loss": 0.3363, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.4187604721399877e-05, "epoch": 1.157556270096463, "percentage": 23.15, "elapsed_time": "0:03:34", "remaining_time": "0:11:53"}
+{"current_steps": 370, "total_steps": 1555, "loss": 0.3453, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.3855612605216174e-05, "epoch": 1.189710610932476, "percentage": 23.79, "elapsed_time": "0:03:40", "remaining_time": "0:11:46"}
+{"current_steps": 380, "total_steps": 1555, "loss": 0.3297, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.351572264645366e-05, "epoch": 1.2218649517684887, "percentage": 24.44, "elapsed_time": "0:03:46", "remaining_time": "0:11:39"}
+{"current_steps": 390, "total_steps": 1555, "loss": 0.3453, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.3168077211072263e-05, "epoch": 1.2540192926045015, "percentage": 25.08, "elapsed_time": "0:03:51", "remaining_time": "0:11:31"}
+{"current_steps": 400, "total_steps": 1555, "loss": 0.34, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.281282191348289e-05, "epoch": 1.2861736334405145, "percentage": 25.72, "elapsed_time": "0:03:57", "remaining_time": "0:11:25"}
+{"current_steps": 400, "total_steps": 1555, "loss": null, "eval_loss": 0.32504987716674805, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 1.2861736334405145, "percentage": 25.72, "elapsed_time": "0:03:57", "remaining_time": "0:11:25"}
+{"current_steps": 410, "total_steps": 1555, "loss": 0.3324, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.245010555555554e-05, "epoch": 1.3183279742765273, "percentage": 26.37, "elapsed_time": "0:04:06", "remaining_time": "0:11:28"}
+{"current_steps": 420, "total_steps": 1555, "loss": 0.3564, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.2080080064292304e-05, "epoch": 1.3504823151125402, "percentage": 27.01, "elapsed_time": "0:04:12", "remaining_time": "0:11:21"}
+{"current_steps": 430, "total_steps": 1555, "loss": 0.3105, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.1702900428191374e-05, "epoch": 1.382636655948553, "percentage": 27.65, "elapsed_time": "0:04:17", "remaining_time": "0:11:14"}
+{"current_steps": 440, "total_steps": 1555, "loss": 0.2817, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.131872463232872e-05, "epoch": 1.414790996784566, "percentage": 28.3, "elapsed_time": "0:04:23", "remaining_time": "0:11:07"}
+{"current_steps": 450, "total_steps": 1555, "loss": 0.3865, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.092771359218461e-05, "epoch": 1.4469453376205788, "percentage": 28.94, "elapsed_time": "0:04:29", "remaining_time": "0:11:00"}
+{"current_steps": 460, "total_steps": 1555, "loss": 0.3358, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.053003108624276e-05, "epoch": 1.4790996784565915, "percentage": 29.58, "elapsed_time": "0:04:34", "remaining_time": "0:10:54"}
+{"current_steps": 470, "total_steps": 1555, "loss": 0.3586, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.0125843687390166e-05, "epoch": 1.5112540192926045, "percentage": 30.23, "elapsed_time": "0:04:40", "remaining_time": "0:10:47"}
+{"current_steps": 480, "total_steps": 1555, "loss": 0.3285, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.9715320693146655e-05, "epoch": 1.5434083601286175, "percentage": 30.87, "elapsed_time": "0:04:46", "remaining_time": "0:10:41"}
+{"current_steps": 490, "total_steps": 1555, "loss": 0.3437, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.9298634054753026e-05, "epoch": 1.5755627009646302, "percentage": 31.51, "elapsed_time": "0:04:51", "remaining_time": "0:10:34"}
+{"current_steps": 500, "total_steps": 1555, "loss": 0.3221, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.887595830514775e-05, "epoch": 1.607717041800643, "percentage": 32.15, "elapsed_time": "0:04:57", "remaining_time": "0:10:26"}
+{"current_steps": 500, "total_steps": 1555, "loss": null, "eval_loss": 0.31748995184898376, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 1.607717041800643, "percentage": 32.15, "elapsed_time": "0:04:57", "remaining_time": "0:10:26"}
+{"current_steps": 510, "total_steps": 1555, "loss": 0.3011, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.844747048586228e-05, "epoch": 1.639871382636656, "percentage": 32.8, "elapsed_time": "0:05:06", "remaining_time": "0:10:27"}
+{"current_steps": 520, "total_steps": 1555, "loss": 0.3461, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.801335007286564e-05, "epoch": 1.6720257234726688, "percentage": 33.44, "elapsed_time": "0:05:12", "remaining_time": "0:10:21"}
+{"current_steps": 530, "total_steps": 1555, "loss": 0.3477, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.757377890138927e-05, "epoch": 1.7041800643086815, "percentage": 34.08, "elapsed_time": "0:05:18", "remaining_time": "0:10:15"}
+{"current_steps": 540, "total_steps": 1555, "loss": 0.3507, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.712894108976372e-05, "epoch": 1.7363344051446945, "percentage": 34.73, "elapsed_time": "0:05:23", "remaining_time": "0:10:08"}
+{"current_steps": 550, "total_steps": 1555, "loss": 0.3327, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.667902296229905e-05, "epoch": 1.7684887459807075, "percentage": 35.37, "elapsed_time": "0:05:29", "remaining_time": "0:10:01"}
+{"current_steps": 560, "total_steps": 1555, "loss": 0.3312, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.622421297124122e-05, "epoch": 1.8006430868167203, "percentage": 36.01, "elapsed_time": "0:05:35", "remaining_time": "0:09:55"}
+{"current_steps": 570, "total_steps": 1555, "loss": 0.3659, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.576470161783712e-05, "epoch": 1.832797427652733, "percentage": 36.66, "elapsed_time": "0:05:40", "remaining_time": "0:09:49"}
+{"current_steps": 580, "total_steps": 1555, "loss": 0.3347, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.5300681372541476e-05, "epoch": 1.864951768488746, "percentage": 37.3, "elapsed_time": "0:05:46", "remaining_time": "0:09:42"}
+{"current_steps": 590, "total_steps": 1555, "loss": 0.3146, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.483234659439889e-05, "epoch": 1.897106109324759, "percentage": 37.94, "elapsed_time": "0:05:52", "remaining_time": "0:09:35"}
+{"current_steps": 600, "total_steps": 1555, "loss": 0.3193, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.435989344963471e-05, "epoch": 1.9292604501607717, "percentage": 38.59, "elapsed_time": "0:05:57", "remaining_time": "0:09:29"}
+{"current_steps": 600, "total_steps": 1555, "loss": null, "eval_loss": 0.3193986415863037, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 1.9292604501607717, "percentage": 38.59, "elapsed_time": "0:05:57", "remaining_time": "0:09:29"}
+{"current_steps": 610, "total_steps": 1555, "loss": 0.3494, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.3883519829489154e-05, "epoch": 1.9614147909967845, "percentage": 39.23, "elapsed_time": "0:06:06", "remaining_time": "0:09:27"}
+{"current_steps": 620, "total_steps": 1555, "loss": 0.2765, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.3403425267328715e-05, "epoch": 1.9935691318327975, "percentage": 39.87, "elapsed_time": "0:06:12", "remaining_time": "0:09:21"}
+{"current_steps": 630, "total_steps": 1555, "loss": 0.327, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.2919810855069864e-05, "epoch": 2.0257234726688105, "percentage": 40.51, "elapsed_time": "0:06:18", "remaining_time": "0:09:15"}
+{"current_steps": 640, "total_steps": 1555, "loss": 0.362, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.243287915894987e-05, "epoch": 2.057877813504823, "percentage": 41.16, "elapsed_time": "0:06:23", "remaining_time": "0:09:08"}
+{"current_steps": 650, "total_steps": 1555, "loss": 0.3393, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.1942834134680124e-05, "epoch": 2.090032154340836, "percentage": 41.8, "elapsed_time": "0:06:29", "remaining_time": "0:09:02"}
+{"current_steps": 660, "total_steps": 1555, "loss": 0.3322, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.144988104201745e-05, "epoch": 2.122186495176849, "percentage": 42.44, "elapsed_time": "0:06:35", "remaining_time": "0:08:55"}
+{"current_steps": 670, "total_steps": 1555, "loss": 0.3191, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.095422635878923e-05, "epoch": 2.154340836012862, "percentage": 43.09, "elapsed_time": "0:06:40", "remaining_time": "0:08:49"}
+{"current_steps": 680, "total_steps": 1555, "loss": 0.3311, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.045607769440829e-05, "epoch": 2.1864951768488745, "percentage": 43.73, "elapsed_time": "0:06:46", "remaining_time": "0:08:43"}
+{"current_steps": 690, "total_steps": 1555, "loss": 0.3412, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.9955643702913865e-05, "epoch": 2.2186495176848875, "percentage": 44.37, "elapsed_time": "0:06:52", "remaining_time": "0:08:36"}
+{"current_steps": 700, "total_steps": 1555, "loss": 0.339, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.9453133995574955e-05, "epoch": 2.2508038585209005, "percentage": 45.02, "elapsed_time": "0:06:57", "remaining_time": "0:08:30"}
+{"current_steps": 700, "total_steps": 1555, "loss": null, "eval_loss": 0.32114845514297485, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 2.2508038585209005, "percentage": 45.02, "elapsed_time": "0:06:57", "remaining_time": "0:08:30"}
+{"current_steps": 710, "total_steps": 1555, "loss": 0.3128, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.8948759053092754e-05, "epoch": 2.282958199356913, "percentage": 45.66, "elapsed_time": "0:07:06", "remaining_time": "0:08:27"}
+{"current_steps": 720, "total_steps": 1555, "loss": 0.3448, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.844273013743896e-05, "epoch": 2.315112540192926, "percentage": 46.3, "elapsed_time": "0:07:12", "remaining_time": "0:08:21"}
+{"current_steps": 730, "total_steps": 1555, "loss": 0.3295, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.7935259203366777e-05, "epoch": 2.347266881028939, "percentage": 46.95, "elapsed_time": "0:07:18", "remaining_time": "0:08:15"}
+{"current_steps": 740, "total_steps": 1555, "loss": 0.3249, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.7426558809631748e-05, "epoch": 2.379421221864952, "percentage": 47.59, "elapsed_time": "0:07:23", "remaining_time": "0:08:08"}
+{"current_steps": 750, "total_steps": 1555, "loss": 0.3213, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.691684202995966e-05, "epoch": 2.4115755627009645, "percentage": 48.23, "elapsed_time": "0:07:29", "remaining_time": "0:08:02"}
+{"current_steps": 760, "total_steps": 1555, "loss": 0.3157, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.6406322363798657e-05, "epoch": 2.4437299035369775, "percentage": 48.87, "elapsed_time": "0:07:35", "remaining_time": "0:07:56"}
+{"current_steps": 770, "total_steps": 1555, "loss": 0.3537, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.589521364689308e-05, "epoch": 2.4758842443729905, "percentage": 49.52, "elapsed_time": "0:07:40", "remaining_time": "0:07:49"}
+{"current_steps": 780, "total_steps": 1555, "loss": 0.3344, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.5383729961716483e-05, "epoch": 2.508038585209003, "percentage": 50.16, "elapsed_time": "0:07:46", "remaining_time": "0:07:43"}
+{"current_steps": 790, "total_steps": 1555, "loss": 0.3398, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.487208554780125e-05, "epoch": 2.540192926045016, "percentage": 50.8, "elapsed_time": "0:07:51", "remaining_time": "0:07:36"}
+{"current_steps": 800, "total_steps": 1555, "loss": 0.3347, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.4360494712002423e-05, "epoch": 2.572347266881029, "percentage": 51.45, "elapsed_time": "0:07:57", "remaining_time": "0:07:30"}
+{"current_steps": 800, "total_steps": 1555, "loss": null, "eval_loss": 0.3262169659137726, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 2.572347266881029, "percentage": 51.45, "elapsed_time": "0:07:57", "remaining_time": "0:07:30"}
+{"current_steps": 800, "total_steps": 1555, "loss": null, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 2.572347266881029, "percentage": 51.45, "elapsed_time": "0:07:57", "remaining_time": "0:07:30"}
+{"current_steps": 17, "total_steps": 17, "loss": null, "eval_loss": 0.31748995184898376, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 2.572347266881029, "percentage": 100.0, "elapsed_time": "0:08:12", "remaining_time": "0:00:00"}
diff --git a/llama2_13b_peft/play_dialog_same_or_different/trainer_state.json b/llama2_13b_peft/play_dialog_same_or_different/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..01b365daf868cd2bea12e2b9d79d841c8dea3962
--- /dev/null
+++ b/llama2_13b_peft/play_dialog_same_or_different/trainer_state.json
@@ -0,0 +1,654 @@
+{
+ "best_metric": 0.31748995184898376,
+ "best_model_checkpoint": "ckpt/llama2_13b_fuze27_no_sys/play_dialog_same_or_different_no_sys/checkpoint-500",
+ "epoch": 2.572347266881029,
+ "eval_steps": 100,
+ "global_step": 800,
+ "is_hyper_param_search": false,
+ "is_local_process_zero": true,
+ "is_world_process_zero": true,
+ "log_history": [
+ {
+ "epoch": 0.03215434083601286,
+ "grad_norm": 11.316423416137695,
+ "learning_rate": 2.5e-05,
+ "loss": 10.7274,
+ "step": 10
+ },
+ {
+ "epoch": 0.06430868167202572,
+ "grad_norm": 13.179275512695312,
+ "learning_rate": 5e-05,
+ "loss": 7.7527,
+ "step": 20
+ },
+ {
+ "epoch": 0.09646302250803858,
+ "grad_norm": 2.945481777191162,
+ "learning_rate": 4.999476426280588e-05,
+ "loss": 1.1271,
+ "step": 30
+ },
+ {
+ "epoch": 0.12861736334405144,
+ "grad_norm": 0.8495484590530396,
+ "learning_rate": 4.997905924425903e-05,
+ "loss": 0.4741,
+ "step": 40
+ },
+ {
+ "epoch": 0.1607717041800643,
+ "grad_norm": 2.0995168685913086,
+ "learning_rate": 4.995289152254744e-05,
+ "loss": 0.3836,
+ "step": 50
+ },
+ {
+ "epoch": 0.19292604501607716,
+ "grad_norm": 2.24106764793396,
+ "learning_rate": 4.991627205825621e-05,
+ "loss": 0.3745,
+ "step": 60
+ },
+ {
+ "epoch": 0.22508038585209003,
+ "grad_norm": 2.550299644470215,
+ "learning_rate": 4.986921618977664e-05,
+ "loss": 0.3658,
+ "step": 70
+ },
+ {
+ "epoch": 0.2572347266881029,
+ "grad_norm": 1.3360856771469116,
+ "learning_rate": 4.981174362688158e-05,
+ "loss": 0.346,
+ "step": 80
+ },
+ {
+ "epoch": 0.28938906752411575,
+ "grad_norm": 1.1294679641723633,
+ "learning_rate": 4.974387844246987e-05,
+ "loss": 0.3624,
+ "step": 90
+ },
+ {
+ "epoch": 0.3215434083601286,
+ "grad_norm": 0.6543893218040466,
+ "learning_rate": 4.9665649062483115e-05,
+ "loss": 0.3588,
+ "step": 100
+ },
+ {
+ "epoch": 0.3215434083601286,
+ "eval_loss": 0.3202289044857025,
+ "eval_runtime": 2.9115,
+ "eval_samples_per_second": 44.993,
+ "eval_steps_per_second": 5.839,
+ "step": 100
+ },
+ {
+ "epoch": 0.3536977491961415,
+ "grad_norm": 0.378452867269516,
+ "learning_rate": 4.957708825399927e-05,
+ "loss": 0.341,
+ "step": 110
+ },
+ {
+ "epoch": 0.3858520900321543,
+ "grad_norm": 0.8828644156455994,
+ "learning_rate": 4.9478233111507856e-05,
+ "loss": 0.3741,
+ "step": 120
+ },
+ {
+ "epoch": 0.4180064308681672,
+ "grad_norm": 0.3250252604484558,
+ "learning_rate": 4.936912504137257e-05,
+ "loss": 0.3254,
+ "step": 130
+ },
+ {
+ "epoch": 0.45016077170418006,
+ "grad_norm": 0.7310301661491394,
+ "learning_rate": 4.924980974448791e-05,
+ "loss": 0.3496,
+ "step": 140
+ },
+ {
+ "epoch": 0.48231511254019294,
+ "grad_norm": 1.3139760494232178,
+ "learning_rate": 4.912033719713687e-05,
+ "loss": 0.3569,
+ "step": 150
+ },
+ {
+ "epoch": 0.5144694533762058,
+ "grad_norm": 1.1433072090148926,
+ "learning_rate": 4.8980761630058014e-05,
+ "loss": 0.3066,
+ "step": 160
+ },
+ {
+ "epoch": 0.5466237942122186,
+ "grad_norm": 0.12614063918590546,
+ "learning_rate": 4.883114150573037e-05,
+ "loss": 0.356,
+ "step": 170
+ },
+ {
+ "epoch": 0.5787781350482315,
+ "grad_norm": 1.8395532369613647,
+ "learning_rate": 4.8671539493885934e-05,
+ "loss": 0.3955,
+ "step": 180
+ },
+ {
+ "epoch": 0.6109324758842444,
+ "grad_norm": 0.9796155691146851,
+ "learning_rate": 4.850202244525987e-05,
+ "loss": 0.3876,
+ "step": 190
+ },
+ {
+ "epoch": 0.6430868167202572,
+ "grad_norm": 0.8637738227844238,
+ "learning_rate": 4.8322661363589507e-05,
+ "loss": 0.321,
+ "step": 200
+ },
+ {
+ "epoch": 0.6430868167202572,
+ "eval_loss": 0.32409238815307617,
+ "eval_runtime": 2.9231,
+ "eval_samples_per_second": 44.816,
+ "eval_steps_per_second": 5.816,
+ "step": 200
+ },
+ {
+ "epoch": 0.6752411575562701,
+ "grad_norm": 0.38207629323005676,
+ "learning_rate": 4.813353137587377e-05,
+ "loss": 0.39,
+ "step": 210
+ },
+ {
+ "epoch": 0.707395498392283,
+ "grad_norm": 0.8289644718170166,
+ "learning_rate": 4.793471170090555e-05,
+ "loss": 0.3429,
+ "step": 220
+ },
+ {
+ "epoch": 0.7395498392282959,
+ "grad_norm": 0.26011592149734497,
+ "learning_rate": 4.772628561609022e-05,
+ "loss": 0.3208,
+ "step": 230
+ },
+ {
+ "epoch": 0.7717041800643086,
+ "grad_norm": 0.571058988571167,
+ "learning_rate": 4.7508340422564134e-05,
+ "loss": 0.344,
+ "step": 240
+ },
+ {
+ "epoch": 0.8038585209003215,
+ "grad_norm": 0.8589864373207092,
+ "learning_rate": 4.728096740862778e-05,
+ "loss": 0.2976,
+ "step": 250
+ },
+ {
+ "epoch": 0.8360128617363344,
+ "grad_norm": 0.08967006206512451,
+ "learning_rate": 4.704426181150884e-05,
+ "loss": 0.3684,
+ "step": 260
+ },
+ {
+ "epoch": 0.8681672025723473,
+ "grad_norm": 0.5221372246742249,
+ "learning_rate": 4.6798322777471216e-05,
+ "loss": 0.3398,
+ "step": 270
+ },
+ {
+ "epoch": 0.9003215434083601,
+ "grad_norm": 1.0666356086730957,
+ "learning_rate": 4.654325332028676e-05,
+ "loss": 0.3236,
+ "step": 280
+ },
+ {
+ "epoch": 0.932475884244373,
+ "grad_norm": 0.281319797039032,
+ "learning_rate": 4.6279160278086994e-05,
+ "loss": 0.2567,
+ "step": 290
+ },
+ {
+ "epoch": 0.9646302250803859,
+ "grad_norm": 0.15490597486495972,
+ "learning_rate": 4.6006154268613014e-05,
+ "loss": 0.3268,
+ "step": 300
+ },
+ {
+ "epoch": 0.9646302250803859,
+ "eval_loss": 0.3192877471446991,
+ "eval_runtime": 2.9293,
+ "eval_samples_per_second": 44.72,
+ "eval_steps_per_second": 5.803,
+ "step": 300
+ },
+ {
+ "epoch": 0.9967845659163987,
+ "grad_norm": 0.10922815650701523,
+ "learning_rate": 4.572434964288226e-05,
+ "loss": 0.3136,
+ "step": 310
+ },
+ {
+ "epoch": 1.0289389067524115,
+ "grad_norm": 0.5204551219940186,
+ "learning_rate": 4.543386443729157e-05,
+ "loss": 0.3386,
+ "step": 320
+ },
+ {
+ "epoch": 1.0610932475884245,
+ "grad_norm": 0.6060867309570312,
+ "learning_rate": 4.513482032417656e-05,
+ "loss": 0.3385,
+ "step": 330
+ },
+ {
+ "epoch": 1.0932475884244373,
+ "grad_norm": 0.8849338889122009,
+ "learning_rate": 4.48273425608481e-05,
+ "loss": 0.3345,
+ "step": 340
+ },
+ {
+ "epoch": 1.1254019292604502,
+ "grad_norm": 0.22567600011825562,
+ "learning_rate": 4.451155993712711e-05,
+ "loss": 0.3515,
+ "step": 350
+ },
+ {
+ "epoch": 1.157556270096463,
+ "grad_norm": 0.1096111610531807,
+ "learning_rate": 4.4187604721399877e-05,
+ "loss": 0.3363,
+ "step": 360
+ },
+ {
+ "epoch": 1.189710610932476,
+ "grad_norm": 0.1326136738061905,
+ "learning_rate": 4.3855612605216174e-05,
+ "loss": 0.3453,
+ "step": 370
+ },
+ {
+ "epoch": 1.2218649517684887,
+ "grad_norm": 0.10588031262159348,
+ "learning_rate": 4.351572264645366e-05,
+ "loss": 0.3297,
+ "step": 380
+ },
+ {
+ "epoch": 1.2540192926045015,
+ "grad_norm": 0.1992698311805725,
+ "learning_rate": 4.3168077211072263e-05,
+ "loss": 0.3453,
+ "step": 390
+ },
+ {
+ "epoch": 1.2861736334405145,
+ "grad_norm": 0.40842464566230774,
+ "learning_rate": 4.281282191348289e-05,
+ "loss": 0.34,
+ "step": 400
+ },
+ {
+ "epoch": 1.2861736334405145,
+ "eval_loss": 0.32504987716674805,
+ "eval_runtime": 2.9329,
+ "eval_samples_per_second": 44.666,
+ "eval_steps_per_second": 5.796,
+ "step": 400
+ },
+ {
+ "epoch": 1.3183279742765273,
+ "grad_norm": 0.34837374091148376,
+ "learning_rate": 4.245010555555554e-05,
+ "loss": 0.3324,
+ "step": 410
+ },
+ {
+ "epoch": 1.3504823151125402,
+ "grad_norm": 0.16926288604736328,
+ "learning_rate": 4.2080080064292304e-05,
+ "loss": 0.3564,
+ "step": 420
+ },
+ {
+ "epoch": 1.382636655948553,
+ "grad_norm": 0.2140832394361496,
+ "learning_rate": 4.1702900428191374e-05,
+ "loss": 0.3105,
+ "step": 430
+ },
+ {
+ "epoch": 1.414790996784566,
+ "grad_norm": 1.07697594165802,
+ "learning_rate": 4.131872463232872e-05,
+ "loss": 0.2817,
+ "step": 440
+ },
+ {
+ "epoch": 1.4469453376205788,
+ "grad_norm": 0.19250230491161346,
+ "learning_rate": 4.092771359218461e-05,
+ "loss": 0.3865,
+ "step": 450
+ },
+ {
+ "epoch": 1.4790996784565915,
+ "grad_norm": 1.0127387046813965,
+ "learning_rate": 4.053003108624276e-05,
+ "loss": 0.3358,
+ "step": 460
+ },
+ {
+ "epoch": 1.5112540192926045,
+ "grad_norm": 0.3083811402320862,
+ "learning_rate": 4.0125843687390166e-05,
+ "loss": 0.3586,
+ "step": 470
+ },
+ {
+ "epoch": 1.5434083601286175,
+ "grad_norm": 0.6321075558662415,
+ "learning_rate": 3.9715320693146655e-05,
+ "loss": 0.3285,
+ "step": 480
+ },
+ {
+ "epoch": 1.5755627009646302,
+ "grad_norm": 0.1519564539194107,
+ "learning_rate": 3.9298634054753026e-05,
+ "loss": 0.3437,
+ "step": 490
+ },
+ {
+ "epoch": 1.607717041800643,
+ "grad_norm": 0.1895354837179184,
+ "learning_rate": 3.887595830514775e-05,
+ "loss": 0.3221,
+ "step": 500
+ },
+ {
+ "epoch": 1.607717041800643,
+ "eval_loss": 0.31748995184898376,
+ "eval_runtime": 2.9296,
+ "eval_samples_per_second": 44.716,
+ "eval_steps_per_second": 5.803,
+ "step": 500
+ },
+ {
+ "epoch": 1.639871382636656,
+ "grad_norm": 0.10975853353738785,
+ "learning_rate": 3.844747048586228e-05,
+ "loss": 0.3011,
+ "step": 510
+ },
+ {
+ "epoch": 1.6720257234726688,
+ "grad_norm": 0.6922447681427002,
+ "learning_rate": 3.801335007286564e-05,
+ "loss": 0.3461,
+ "step": 520
+ },
+ {
+ "epoch": 1.7041800643086815,
+ "grad_norm": 0.40238264203071594,
+ "learning_rate": 3.757377890138927e-05,
+ "loss": 0.3477,
+ "step": 530
+ },
+ {
+ "epoch": 1.7363344051446945,
+ "grad_norm": 0.5762842297554016,
+ "learning_rate": 3.712894108976372e-05,
+ "loss": 0.3507,
+ "step": 540
+ },
+ {
+ "epoch": 1.7684887459807075,
+ "grad_norm": 0.5789851546287537,
+ "learning_rate": 3.667902296229905e-05,
+ "loss": 0.3327,
+ "step": 550
+ },
+ {
+ "epoch": 1.8006430868167203,
+ "grad_norm": 1.1416003704071045,
+ "learning_rate": 3.622421297124122e-05,
+ "loss": 0.3312,
+ "step": 560
+ },
+ {
+ "epoch": 1.832797427652733,
+ "grad_norm": 0.2402690201997757,
+ "learning_rate": 3.576470161783712e-05,
+ "loss": 0.3659,
+ "step": 570
+ },
+ {
+ "epoch": 1.864951768488746,
+ "grad_norm": 0.17154522240161896,
+ "learning_rate": 3.5300681372541476e-05,
+ "loss": 0.3347,
+ "step": 580
+ },
+ {
+ "epoch": 1.897106109324759,
+ "grad_norm": 0.14049768447875977,
+ "learning_rate": 3.483234659439889e-05,
+ "loss": 0.3146,
+ "step": 590
+ },
+ {
+ "epoch": 1.9292604501607717,
+ "grad_norm": 0.23624245822429657,
+ "learning_rate": 3.435989344963471e-05,
+ "loss": 0.3193,
+ "step": 600
+ },
+ {
+ "epoch": 1.9292604501607717,
+ "eval_loss": 0.3193986415863037,
+ "eval_runtime": 2.9308,
+ "eval_samples_per_second": 44.698,
+ "eval_steps_per_second": 5.801,
+ "step": 600
+ },
+ {
+ "epoch": 1.9614147909967845,
+ "grad_norm": 0.11554815620183945,
+ "learning_rate": 3.3883519829489154e-05,
+ "loss": 0.3494,
+ "step": 610
+ },
+ {
+ "epoch": 1.9935691318327975,
+ "grad_norm": 0.15566588938236237,
+ "learning_rate": 3.3403425267328715e-05,
+ "loss": 0.2765,
+ "step": 620
+ },
+ {
+ "epoch": 2.0257234726688105,
+ "grad_norm": 0.10102688521146774,
+ "learning_rate": 3.2919810855069864e-05,
+ "loss": 0.327,
+ "step": 630
+ },
+ {
+ "epoch": 2.057877813504823,
+ "grad_norm": 0.26385530829429626,
+ "learning_rate": 3.243287915894987e-05,
+ "loss": 0.362,
+ "step": 640
+ },
+ {
+ "epoch": 2.090032154340836,
+ "grad_norm": 0.14168129861354828,
+ "learning_rate": 3.1942834134680124e-05,
+ "loss": 0.3393,
+ "step": 650
+ },
+ {
+ "epoch": 2.122186495176849,
+ "grad_norm": 0.10720091313123703,
+ "learning_rate": 3.144988104201745e-05,
+ "loss": 0.3322,
+ "step": 660
+ },
+ {
+ "epoch": 2.154340836012862,
+ "grad_norm": 0.12918303906917572,
+ "learning_rate": 3.095422635878923e-05,
+ "loss": 0.3191,
+ "step": 670
+ },
+ {
+ "epoch": 2.1864951768488745,
+ "grad_norm": 0.13608509302139282,
+ "learning_rate": 3.045607769440829e-05,
+ "loss": 0.3311,
+ "step": 680
+ },
+ {
+ "epoch": 2.2186495176848875,
+ "grad_norm": 0.2435235232114792,
+ "learning_rate": 2.9955643702913865e-05,
+ "loss": 0.3412,
+ "step": 690
+ },
+ {
+ "epoch": 2.2508038585209005,
+ "grad_norm": 0.4761883616447449,
+ "learning_rate": 2.9453133995574955e-05,
+ "loss": 0.339,
+ "step": 700
+ },
+ {
+ "epoch": 2.2508038585209005,
+ "eval_loss": 0.32114845514297485,
+ "eval_runtime": 2.9341,
+ "eval_samples_per_second": 44.647,
+ "eval_steps_per_second": 5.794,
+ "step": 700
+ },
+ {
+ "epoch": 2.282958199356913,
+ "grad_norm": 0.47587987780570984,
+ "learning_rate": 2.8948759053092754e-05,
+ "loss": 0.3128,
+ "step": 710
+ },
+ {
+ "epoch": 2.315112540192926,
+ "grad_norm": 0.11377859860658646,
+ "learning_rate": 2.844273013743896e-05,
+ "loss": 0.3448,
+ "step": 720
+ },
+ {
+ "epoch": 2.347266881028939,
+ "grad_norm": 0.7121096253395081,
+ "learning_rate": 2.7935259203366777e-05,
+ "loss": 0.3295,
+ "step": 730
+ },
+ {
+ "epoch": 2.379421221864952,
+ "grad_norm": 0.6000668406486511,
+ "learning_rate": 2.7426558809631748e-05,
+ "loss": 0.3249,
+ "step": 740
+ },
+ {
+ "epoch": 2.4115755627009645,
+ "grad_norm": 0.14565663039684296,
+ "learning_rate": 2.691684202995966e-05,
+ "loss": 0.3213,
+ "step": 750
+ },
+ {
+ "epoch": 2.4437299035369775,
+ "grad_norm": 0.12155289202928543,
+ "learning_rate": 2.6406322363798657e-05,
+ "loss": 0.3157,
+ "step": 760
+ },
+ {
+ "epoch": 2.4758842443729905,
+ "grad_norm": 1.0030161142349243,
+ "learning_rate": 2.589521364689308e-05,
+ "loss": 0.3537,
+ "step": 770
+ },
+ {
+ "epoch": 2.508038585209003,
+ "grad_norm": 0.19715431332588196,
+ "learning_rate": 2.5383729961716483e-05,
+ "loss": 0.3344,
+ "step": 780
+ },
+ {
+ "epoch": 2.540192926045016,
+ "grad_norm": 0.15441320836544037,
+ "learning_rate": 2.487208554780125e-05,
+ "loss": 0.3398,
+ "step": 790
+ },
+ {
+ "epoch": 2.572347266881029,
+ "grad_norm": 0.22851267457008362,
+ "learning_rate": 2.4360494712002423e-05,
+ "loss": 0.3347,
+ "step": 800
+ },
+ {
+ "epoch": 2.572347266881029,
+ "eval_loss": 0.3262169659137726,
+ "eval_runtime": 2.9319,
+ "eval_samples_per_second": 44.681,
+ "eval_steps_per_second": 5.798,
+ "step": 800
+ },
+ {
+ "epoch": 2.572347266881029,
+ "step": 800,
+ "total_flos": 1.4161342881595392e+17,
+ "train_loss": 0.5724774128198624,
+ "train_runtime": 490.2786,
+ "train_samples_per_second": 25.302,
+ "train_steps_per_second": 3.172
+ }
+ ],
+ "logging_steps": 10,
+ "max_steps": 1555,
+ "num_input_tokens_seen": 0,
+ "num_train_epochs": 5,
+ "save_steps": 100,
+ "total_flos": 1.4161342881595392e+17,
+ "train_batch_size": 4,
+ "trial_name": null,
+ "trial_params": null
+}
diff --git a/llama2_13b_peft/play_dialog_same_or_different/training_args.bin b/llama2_13b_peft/play_dialog_same_or_different/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..f1d1f733358c2532d2f9886ed5a204e717660c7a
--- /dev/null
+++ b/llama2_13b_peft/play_dialog_same_or_different/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cd8f472354811e14472287f3112efe50d2e4c9ca31cc439047b7e79f5de56f64
+size 5240
diff --git a/llama2_13b_peft/play_dialog_same_or_different/training_eval_loss.png b/llama2_13b_peft/play_dialog_same_or_different/training_eval_loss.png
new file mode 100644
index 0000000000000000000000000000000000000000..e5e71a8280e6b4ac125a6cbf982eb5e5cce1ecb7
Binary files /dev/null and b/llama2_13b_peft/play_dialog_same_or_different/training_eval_loss.png differ
diff --git a/llama2_13b_peft/play_dialog_same_or_different/training_loss.png b/llama2_13b_peft/play_dialog_same_or_different/training_loss.png
new file mode 100644
index 0000000000000000000000000000000000000000..6c126efbd85450de53669bbacf011d8d77b9f56f
Binary files /dev/null and b/llama2_13b_peft/play_dialog_same_or_different/training_loss.png differ
diff --git a/llama2_13b_peft/question_selection/README.md b/llama2_13b_peft/question_selection/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..03bc4c9ae76f1036a042285f8e8bb71e62b10c37
--- /dev/null
+++ b/llama2_13b_peft/question_selection/README.md
@@ -0,0 +1,69 @@
+---
+license: other
+library_name: peft
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+base_model: /data1/model/llama2/meta-llama/Llama2-13b
+model-index:
+- name: question_selection_no_sys
+ results: []
+---
+
+
+
+# question_selection_no_sys
+
+This model is a fine-tuned version of [/data1/model/llama2/meta-llama/Llama2-13b](https://huggingface.co//data1/model/llama2/meta-llama/Llama2-13b) on the question_selection_no_sys dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.0145
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0005
+- train_batch_size: 8
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 3
+- total_train_batch_size: 24
+- total_eval_batch_size: 24
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_steps: 20
+- num_epochs: 10.0
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 0.013 | 2.22 | 100 | 0.0138 |
+| 0.0034 | 4.44 | 200 | 0.0183 |
+| 0.0001 | 6.67 | 300 | 0.0147 |
+| 0.0 | 8.89 | 400 | 0.0143 |
+
+
+### Framework versions
+
+- PEFT 0.9.0
+- Transformers 4.38.2
+- Pytorch 2.2.1
+- Datasets 2.18.0
+- Tokenizers 0.15.2
\ No newline at end of file
diff --git a/llama2_13b_peft/question_selection/adapter_config.json b/llama2_13b_peft/question_selection/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..90b1c20c2ec0aec78456689058f96026625ad9d6
--- /dev/null
+++ b/llama2_13b_peft/question_selection/adapter_config.json
@@ -0,0 +1,33 @@
+{
+ "alpha_pattern": {},
+ "auto_mapping": null,
+ "base_model_name_or_path": "/data1/model/llama2/meta-llama/Llama2-13b",
+ "bias": "none",
+ "fan_in_fan_out": false,
+ "inference_mode": true,
+ "init_lora_weights": true,
+ "layers_pattern": null,
+ "layers_to_transform": null,
+ "loftq_config": {},
+ "lora_alpha": 16,
+ "lora_dropout": 0.0,
+ "megatron_config": null,
+ "megatron_core": "megatron.core",
+ "modules_to_save": null,
+ "peft_type": "LORA",
+ "r": 8,
+ "rank_pattern": {},
+ "revision": null,
+ "target_modules": [
+ "q_proj",
+ "v_proj",
+ "down_proj",
+ "gate_proj",
+ "k_proj",
+ "o_proj",
+ "up_proj"
+ ],
+ "task_type": "CAUSAL_LM",
+ "use_dora": false,
+ "use_rslora": false
+}
\ No newline at end of file
diff --git a/llama2_13b_peft/question_selection/adapter_model.safetensors b/llama2_13b_peft/question_selection/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..e7d514fa6137ad0339f4fee489d14d65673deecd
--- /dev/null
+++ b/llama2_13b_peft/question_selection/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14b62eee50bb68fb7f84fecf1df8e492a6fa0ce84af109e4a47f551691f19426
+size 125248064
diff --git a/llama2_13b_peft/question_selection/all_results.json b/llama2_13b_peft/question_selection/all_results.json
new file mode 100644
index 0000000000000000000000000000000000000000..3993178e7ecf27f202949bfae30e333625e82b62
--- /dev/null
+++ b/llama2_13b_peft/question_selection/all_results.json
@@ -0,0 +1,11 @@
+{
+ "epoch": 10.0,
+ "eval_loss": 0.014549500308930874,
+ "eval_runtime": 4.8285,
+ "eval_samples_per_second": 39.349,
+ "eval_steps_per_second": 1.657,
+ "train_loss": 0.025018121934683425,
+ "train_runtime": 909.2296,
+ "train_samples_per_second": 11.834,
+ "train_steps_per_second": 0.495
+}
\ No newline at end of file
diff --git a/llama2_13b_peft/question_selection/eval_results.json b/llama2_13b_peft/question_selection/eval_results.json
new file mode 100644
index 0000000000000000000000000000000000000000..06fadb1112ee7df3901a72dca203ce6a89f639d8
--- /dev/null
+++ b/llama2_13b_peft/question_selection/eval_results.json
@@ -0,0 +1,7 @@
+{
+ "epoch": 10.0,
+ "eval_loss": 0.014549500308930874,
+ "eval_runtime": 4.8285,
+ "eval_samples_per_second": 39.349,
+ "eval_steps_per_second": 1.657
+}
\ No newline at end of file
diff --git a/llama2_13b_peft/question_selection/special_tokens_map.json b/llama2_13b_peft/question_selection/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..72ecfeeb7e14d244c936169d2ed139eeae235ef1
--- /dev/null
+++ b/llama2_13b_peft/question_selection/special_tokens_map.json
@@ -0,0 +1,24 @@
+{
+ "bos_token": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false
+ },
+ "eos_token": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false
+ },
+ "pad_token": "",
+ "unk_token": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false
+ }
+}
diff --git a/llama2_13b_peft/question_selection/tokenizer.model b/llama2_13b_peft/question_selection/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..6c00c742ce03c627d6cd5b795984876fa49fa899
--- /dev/null
+++ b/llama2_13b_peft/question_selection/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723
diff --git a/llama2_13b_peft/question_selection/tokenizer_config.json b/llama2_13b_peft/question_selection/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..a40266f39e5b5fed14de34710d35eb9e98d6bdad
--- /dev/null
+++ b/llama2_13b_peft/question_selection/tokenizer_config.json
@@ -0,0 +1,45 @@
+{
+ "add_bos_token": true,
+ "add_eos_token": false,
+ "add_prefix_space": true,
+ "added_tokens_decoder": {
+ "0": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false,
+ "special": true
+ },
+ "1": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false,
+ "special": true
+ },
+ "2": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false,
+ "special": true
+ }
+ },
+ "bos_token": "",
+ "chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '\\n' }}{% endif %}{% endfor %}",
+ "clean_up_tokenization_spaces": false,
+ "eos_token": "",
+ "legacy": true,
+ "model_max_length": 1000000000000000019884624838656,
+ "pad_token": "",
+ "padding_side": "right",
+ "sp_model_kwargs": {},
+ "spaces_between_special_tokens": false,
+ "split_special_tokens": false,
+ "tokenizer_class": "LlamaTokenizer",
+ "unk_token": "",
+ "use_default_system_prompt": false
+}
diff --git a/llama2_13b_peft/question_selection/train_results.json b/llama2_13b_peft/question_selection/train_results.json
new file mode 100644
index 0000000000000000000000000000000000000000..28c2c82dca4773888e7c4f28f325e0f2d912d3a6
--- /dev/null
+++ b/llama2_13b_peft/question_selection/train_results.json
@@ -0,0 +1,7 @@
+{
+ "epoch": 10.0,
+ "train_loss": 0.025018121934683425,
+ "train_runtime": 909.2296,
+ "train_samples_per_second": 11.834,
+ "train_steps_per_second": 0.495
+}
\ No newline at end of file
diff --git a/llama2_13b_peft/question_selection/trainer_log.jsonl b/llama2_13b_peft/question_selection/trainer_log.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..2a73afae570719402ed25b2ce5728f3e1f936fc9
--- /dev/null
+++ b/llama2_13b_peft/question_selection/trainer_log.jsonl
@@ -0,0 +1,51 @@
+{"current_steps": 10, "total_steps": 450, "loss": 0.7467, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00025, "epoch": 0.22, "percentage": 2.22, "elapsed_time": "0:00:20", "remaining_time": "0:14:55"}
+{"current_steps": 20, "total_steps": 450, "loss": 0.0676, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0005, "epoch": 0.44, "percentage": 4.44, "elapsed_time": "0:00:42", "remaining_time": "0:15:05"}
+{"current_steps": 30, "total_steps": 450, "loss": 0.0669, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0004993330709158879, "epoch": 0.67, "percentage": 6.67, "elapsed_time": "0:01:00", "remaining_time": "0:14:05"}
+{"current_steps": 40, "total_steps": 450, "loss": 0.0401, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0004973358420187776, "epoch": 0.89, "percentage": 8.89, "elapsed_time": "0:01:20", "remaining_time": "0:13:44"}
+{"current_steps": 50, "total_steps": 450, "loss": 0.0267, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0004940189693889818, "epoch": 1.11, "percentage": 11.11, "elapsed_time": "0:01:40", "remaining_time": "0:13:26"}
+{"current_steps": 60, "total_steps": 450, "loss": 0.0172, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0004894001499771015, "epoch": 1.33, "percentage": 13.33, "elapsed_time": "0:02:00", "remaining_time": "0:13:02"}
+{"current_steps": 70, "total_steps": 450, "loss": 0.0141, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00048350402718313703, "epoch": 1.56, "percentage": 15.56, "elapsed_time": "0:02:18", "remaining_time": "0:12:29"}
+{"current_steps": 80, "total_steps": 450, "loss": 0.0225, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0004763620593732867, "epoch": 1.78, "percentage": 17.78, "elapsed_time": "0:02:37", "remaining_time": "0:12:06"}
+{"current_steps": 90, "total_steps": 450, "loss": 0.0182, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00046801235203595195, "epoch": 2.0, "percentage": 20.0, "elapsed_time": "0:02:57", "remaining_time": "0:11:49"}
+{"current_steps": 100, "total_steps": 450, "loss": 0.013, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0004584994544724695, "epoch": 2.22, "percentage": 22.22, "elapsed_time": "0:03:17", "remaining_time": "0:11:31"}
+{"current_steps": 100, "total_steps": 450, "loss": null, "eval_loss": 0.013783249072730541, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 2.22, "percentage": 22.22, "elapsed_time": "0:03:17", "remaining_time": "0:11:31"}
+{"current_steps": 110, "total_steps": 450, "loss": 0.0078, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00044787412210731353, "epoch": 2.44, "percentage": 24.44, "elapsed_time": "0:03:42", "remaining_time": "0:11:28"}
+{"current_steps": 120, "total_steps": 450, "loss": 0.0104, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00043619304568594545, "epoch": 2.67, "percentage": 26.67, "elapsed_time": "0:04:01", "remaining_time": "0:11:05"}
+{"current_steps": 130, "total_steps": 450, "loss": 0.0191, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0004235185488051585, "epoch": 2.89, "percentage": 28.89, "elapsed_time": "0:04:20", "remaining_time": "0:10:41"}
+{"current_steps": 140, "total_steps": 450, "loss": 0.0065, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0004099182553897228, "epoch": 3.11, "percentage": 31.11, "elapsed_time": "0:04:38", "remaining_time": "0:10:16"}
+{"current_steps": 150, "total_steps": 450, "loss": 0.0034, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00039546472888948826, "epoch": 3.33, "percentage": 33.33, "elapsed_time": "0:04:57", "remaining_time": "0:09:54"}
+{"current_steps": 160, "total_steps": 450, "loss": 0.007, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0003802350851219826, "epoch": 3.56, "percentage": 35.56, "elapsed_time": "0:05:20", "remaining_time": "0:09:40"}
+{"current_steps": 170, "total_steps": 450, "loss": 0.0058, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00036431058082615964, "epoch": 3.78, "percentage": 37.78, "elapsed_time": "0:05:38", "remaining_time": "0:09:18"}
+{"current_steps": 180, "total_steps": 450, "loss": 0.0039, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00034777618012253895, "epoch": 4.0, "percentage": 40.0, "elapsed_time": "0:05:57", "remaining_time": "0:08:56"}
+{"current_steps": 190, "total_steps": 450, "loss": 0.0035, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00033072010119286155, "epoch": 4.22, "percentage": 42.22, "elapsed_time": "0:06:16", "remaining_time": "0:08:35"}
+{"current_steps": 200, "total_steps": 450, "loss": 0.0034, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0003132333455979202, "epoch": 4.44, "percentage": 44.44, "elapsed_time": "0:06:37", "remaining_time": "0:08:17"}
+{"current_steps": 200, "total_steps": 450, "loss": null, "eval_loss": 0.01827273517847061, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 4.44, "percentage": 44.44, "elapsed_time": "0:06:37", "remaining_time": "0:08:17"}
+{"current_steps": 210, "total_steps": 450, "loss": 0.0043, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0002954092127448591, "epoch": 4.67, "percentage": 46.67, "elapsed_time": "0:07:02", "remaining_time": "0:08:03"}
+{"current_steps": 220, "total_steps": 450, "loss": 0.0033, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00027734280209446867, "epoch": 4.89, "percentage": 48.89, "elapsed_time": "0:07:22", "remaining_time": "0:07:42"}
+{"current_steps": 230, "total_steps": 450, "loss": 0.0051, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00025913050576441477, "epoch": 5.11, "percentage": 51.11, "elapsed_time": "0:07:40", "remaining_time": "0:07:20"}
+{"current_steps": 240, "total_steps": 450, "loss": 0.0011, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00024086949423558527, "epoch": 5.33, "percentage": 53.33, "elapsed_time": "0:07:59", "remaining_time": "0:06:59"}
+{"current_steps": 250, "total_steps": 450, "loss": 0.0032, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00022265719790553147, "epoch": 5.56, "percentage": 55.56, "elapsed_time": "0:08:20", "remaining_time": "0:06:40"}
+{"current_steps": 260, "total_steps": 450, "loss": 0.0016, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0002045907872551409, "epoch": 5.78, "percentage": 57.78, "elapsed_time": "0:08:41", "remaining_time": "0:06:20"}
+{"current_steps": 270, "total_steps": 450, "loss": 0.0008, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0001867666544020798, "epoch": 6.0, "percentage": 60.0, "elapsed_time": "0:09:00", "remaining_time": "0:06:00"}
+{"current_steps": 280, "total_steps": 450, "loss": 0.0008, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00016927989880713852, "epoch": 6.22, "percentage": 62.22, "elapsed_time": "0:09:22", "remaining_time": "0:05:41"}
+{"current_steps": 290, "total_steps": 450, "loss": 0.0002, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00015222381987746103, "epoch": 6.44, "percentage": 64.44, "elapsed_time": "0:09:41", "remaining_time": "0:05:20"}
+{"current_steps": 300, "total_steps": 450, "loss": 0.0001, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00013568941917384037, "epoch": 6.67, "percentage": 66.67, "elapsed_time": "0:09:58", "remaining_time": "0:04:59"}
+{"current_steps": 300, "total_steps": 450, "loss": null, "eval_loss": 0.01472838968038559, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 6.67, "percentage": 66.67, "elapsed_time": "0:09:58", "remaining_time": "0:04:59"}
+{"current_steps": 310, "total_steps": 450, "loss": 0.0008, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00011976491487801746, "epoch": 6.89, "percentage": 68.89, "elapsed_time": "0:10:22", "remaining_time": "0:04:41"}
+{"current_steps": 320, "total_steps": 450, "loss": 0.0002, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.00010453527111051184, "epoch": 7.11, "percentage": 71.11, "elapsed_time": "0:10:42", "remaining_time": "0:04:20"}
+{"current_steps": 330, "total_steps": 450, "loss": 0.0, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.008174461027724e-05, "epoch": 7.33, "percentage": 73.33, "elapsed_time": "0:11:02", "remaining_time": "0:04:00"}
+{"current_steps": 340, "total_steps": 450, "loss": 0.0001, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.648145119484151e-05, "epoch": 7.56, "percentage": 75.56, "elapsed_time": "0:11:22", "remaining_time": "0:03:40"}
+{"current_steps": 350, "total_steps": 450, "loss": 0.0, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.380695431405453e-05, "epoch": 7.78, "percentage": 77.78, "elapsed_time": "0:11:40", "remaining_time": "0:03:20"}
+{"current_steps": 360, "total_steps": 450, "loss": 0.0, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.21258778926865e-05, "epoch": 8.0, "percentage": 80.0, "elapsed_time": "0:12:01", "remaining_time": "0:03:00"}
+{"current_steps": 370, "total_steps": 450, "loss": 0.0, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.150054552753055e-05, "epoch": 8.22, "percentage": 82.22, "elapsed_time": "0:12:19", "remaining_time": "0:02:39"}
+{"current_steps": 380, "total_steps": 450, "loss": 0.0, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.198764796404807e-05, "epoch": 8.44, "percentage": 84.44, "elapsed_time": "0:12:39", "remaining_time": "0:02:19"}
+{"current_steps": 390, "total_steps": 450, "loss": 0.0, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.3637940626713344e-05, "epoch": 8.67, "percentage": 86.67, "elapsed_time": "0:13:00", "remaining_time": "0:02:00"}
+{"current_steps": 400, "total_steps": 450, "loss": 0.0, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.649597281686302e-05, "epoch": 8.89, "percentage": 88.89, "elapsed_time": "0:13:19", "remaining_time": "0:01:39"}
+{"current_steps": 400, "total_steps": 450, "loss": null, "eval_loss": 0.014345706440508366, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 8.89, "percentage": 88.89, "elapsed_time": "0:13:19", "remaining_time": "0:01:39"}
+{"current_steps": 410, "total_steps": 450, "loss": 0.0001, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0599850022898538e-05, "epoch": 9.11, "percentage": 91.11, "elapsed_time": "0:13:44", "remaining_time": "0:01:20"}
+{"current_steps": 420, "total_steps": 450, "loss": 0.0, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.981030611018234e-06, "epoch": 9.33, "percentage": 93.33, "elapsed_time": "0:14:03", "remaining_time": "0:01:00"}
+{"current_steps": 430, "total_steps": 450, "loss": 0.0, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.664157981222437e-06, "epoch": 9.56, "percentage": 95.56, "elapsed_time": "0:14:23", "remaining_time": "0:00:40"}
+{"current_steps": 440, "total_steps": 450, "loss": 0.0, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.66929084112089e-07, "epoch": 9.78, "percentage": 97.78, "elapsed_time": "0:14:42", "remaining_time": "0:00:20"}
+{"current_steps": 450, "total_steps": 450, "loss": 0.0, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0, "epoch": 10.0, "percentage": 100.0, "elapsed_time": "0:15:03", "remaining_time": "0:00:00"}
+{"current_steps": 450, "total_steps": 450, "loss": null, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 10.0, "percentage": 100.0, "elapsed_time": "0:15:03", "remaining_time": "0:00:00"}
+{"current_steps": 8, "total_steps": 8, "loss": null, "eval_loss": 0.014549500308930874, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 10.0, "percentage": 100.0, "elapsed_time": "0:15:08", "remaining_time": "0:00:00"}
diff --git a/llama2_13b_peft/question_selection/trainer_state.json b/llama2_13b_peft/question_selection/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..4474270496dbf4f70aab9c8cf1b5262999b850b1
--- /dev/null
+++ b/llama2_13b_peft/question_selection/trainer_state.json
@@ -0,0 +1,377 @@
+{
+ "best_metric": null,
+ "best_model_checkpoint": null,
+ "epoch": 10.0,
+ "eval_steps": 100,
+ "global_step": 450,
+ "is_hyper_param_search": false,
+ "is_local_process_zero": true,
+ "is_world_process_zero": true,
+ "log_history": [
+ {
+ "epoch": 0.22,
+ "grad_norm": 0.2616782486438751,
+ "learning_rate": 0.00025,
+ "loss": 0.7467,
+ "step": 10
+ },
+ {
+ "epoch": 0.44,
+ "grad_norm": 0.4230063855648041,
+ "learning_rate": 0.0005,
+ "loss": 0.0676,
+ "step": 20
+ },
+ {
+ "epoch": 0.67,
+ "grad_norm": 0.4279615581035614,
+ "learning_rate": 0.0004993330709158879,
+ "loss": 0.0669,
+ "step": 30
+ },
+ {
+ "epoch": 0.89,
+ "grad_norm": 0.26194414496421814,
+ "learning_rate": 0.0004973358420187776,
+ "loss": 0.0401,
+ "step": 40
+ },
+ {
+ "epoch": 1.11,
+ "grad_norm": 0.18304042518138885,
+ "learning_rate": 0.0004940189693889818,
+ "loss": 0.0267,
+ "step": 50
+ },
+ {
+ "epoch": 1.33,
+ "grad_norm": 0.15659470856189728,
+ "learning_rate": 0.0004894001499771015,
+ "loss": 0.0172,
+ "step": 60
+ },
+ {
+ "epoch": 1.56,
+ "grad_norm": 0.8856974840164185,
+ "learning_rate": 0.00048350402718313703,
+ "loss": 0.0141,
+ "step": 70
+ },
+ {
+ "epoch": 1.78,
+ "grad_norm": 0.2954822778701782,
+ "learning_rate": 0.0004763620593732867,
+ "loss": 0.0225,
+ "step": 80
+ },
+ {
+ "epoch": 2.0,
+ "grad_norm": 0.36144667863845825,
+ "learning_rate": 0.00046801235203595195,
+ "loss": 0.0182,
+ "step": 90
+ },
+ {
+ "epoch": 2.22,
+ "grad_norm": 0.20678772032260895,
+ "learning_rate": 0.0004584994544724695,
+ "loss": 0.013,
+ "step": 100
+ },
+ {
+ "epoch": 2.22,
+ "eval_loss": 0.013783249072730541,
+ "eval_runtime": 4.8183,
+ "eval_samples_per_second": 39.433,
+ "eval_steps_per_second": 1.66,
+ "step": 100
+ },
+ {
+ "epoch": 2.44,
+ "grad_norm": 0.3987113833427429,
+ "learning_rate": 0.00044787412210731353,
+ "loss": 0.0078,
+ "step": 110
+ },
+ {
+ "epoch": 2.67,
+ "grad_norm": 0.23118920624256134,
+ "learning_rate": 0.00043619304568594545,
+ "loss": 0.0104,
+ "step": 120
+ },
+ {
+ "epoch": 2.89,
+ "grad_norm": 0.1873694211244583,
+ "learning_rate": 0.0004235185488051585,
+ "loss": 0.0191,
+ "step": 130
+ },
+ {
+ "epoch": 3.11,
+ "grad_norm": 0.03233395144343376,
+ "learning_rate": 0.0004099182553897228,
+ "loss": 0.0065,
+ "step": 140
+ },
+ {
+ "epoch": 3.33,
+ "grad_norm": 0.5932971239089966,
+ "learning_rate": 0.00039546472888948826,
+ "loss": 0.0034,
+ "step": 150
+ },
+ {
+ "epoch": 3.56,
+ "grad_norm": 0.13725461065769196,
+ "learning_rate": 0.0003802350851219826,
+ "loss": 0.007,
+ "step": 160
+ },
+ {
+ "epoch": 3.78,
+ "grad_norm": 0.22745588421821594,
+ "learning_rate": 0.00036431058082615964,
+ "loss": 0.0058,
+ "step": 170
+ },
+ {
+ "epoch": 4.0,
+ "grad_norm": 0.0794651210308075,
+ "learning_rate": 0.00034777618012253895,
+ "loss": 0.0039,
+ "step": 180
+ },
+ {
+ "epoch": 4.22,
+ "grad_norm": 0.06090324744582176,
+ "learning_rate": 0.00033072010119286155,
+ "loss": 0.0035,
+ "step": 190
+ },
+ {
+ "epoch": 4.44,
+ "grad_norm": 0.17350110411643982,
+ "learning_rate": 0.0003132333455979202,
+ "loss": 0.0034,
+ "step": 200
+ },
+ {
+ "epoch": 4.44,
+ "eval_loss": 0.01827273517847061,
+ "eval_runtime": 4.8208,
+ "eval_samples_per_second": 39.413,
+ "eval_steps_per_second": 1.659,
+ "step": 200
+ },
+ {
+ "epoch": 4.67,
+ "grad_norm": 0.0679766833782196,
+ "learning_rate": 0.0002954092127448591,
+ "loss": 0.0043,
+ "step": 210
+ },
+ {
+ "epoch": 4.89,
+ "grad_norm": 0.2287779450416565,
+ "learning_rate": 0.00027734280209446867,
+ "loss": 0.0033,
+ "step": 220
+ },
+ {
+ "epoch": 5.11,
+ "grad_norm": 0.022623347118496895,
+ "learning_rate": 0.00025913050576441477,
+ "loss": 0.0051,
+ "step": 230
+ },
+ {
+ "epoch": 5.33,
+ "grad_norm": 0.023388344794511795,
+ "learning_rate": 0.00024086949423558527,
+ "loss": 0.0011,
+ "step": 240
+ },
+ {
+ "epoch": 5.56,
+ "grad_norm": 0.06873177736997604,
+ "learning_rate": 0.00022265719790553147,
+ "loss": 0.0032,
+ "step": 250
+ },
+ {
+ "epoch": 5.78,
+ "grad_norm": 0.00556823518127203,
+ "learning_rate": 0.0002045907872551409,
+ "loss": 0.0016,
+ "step": 260
+ },
+ {
+ "epoch": 6.0,
+ "grad_norm": 0.00377739779651165,
+ "learning_rate": 0.0001867666544020798,
+ "loss": 0.0008,
+ "step": 270
+ },
+ {
+ "epoch": 6.22,
+ "grad_norm": 0.017272206023335457,
+ "learning_rate": 0.00016927989880713852,
+ "loss": 0.0008,
+ "step": 280
+ },
+ {
+ "epoch": 6.44,
+ "grad_norm": 0.028082353994250298,
+ "learning_rate": 0.00015222381987746103,
+ "loss": 0.0002,
+ "step": 290
+ },
+ {
+ "epoch": 6.67,
+ "grad_norm": 0.0010064858943223953,
+ "learning_rate": 0.00013568941917384037,
+ "loss": 0.0001,
+ "step": 300
+ },
+ {
+ "epoch": 6.67,
+ "eval_loss": 0.01472838968038559,
+ "eval_runtime": 4.8243,
+ "eval_samples_per_second": 39.384,
+ "eval_steps_per_second": 1.658,
+ "step": 300
+ },
+ {
+ "epoch": 6.89,
+ "grad_norm": 0.0007610166212543845,
+ "learning_rate": 0.00011976491487801746,
+ "loss": 0.0008,
+ "step": 310
+ },
+ {
+ "epoch": 7.11,
+ "grad_norm": 0.04695134982466698,
+ "learning_rate": 0.00010453527111051184,
+ "loss": 0.0002,
+ "step": 320
+ },
+ {
+ "epoch": 7.33,
+ "grad_norm": 0.0033624598290771246,
+ "learning_rate": 9.008174461027724e-05,
+ "loss": 0.0,
+ "step": 330
+ },
+ {
+ "epoch": 7.56,
+ "grad_norm": 0.003984487149864435,
+ "learning_rate": 7.648145119484151e-05,
+ "loss": 0.0001,
+ "step": 340
+ },
+ {
+ "epoch": 7.78,
+ "grad_norm": 0.002355258446186781,
+ "learning_rate": 6.380695431405453e-05,
+ "loss": 0.0,
+ "step": 350
+ },
+ {
+ "epoch": 8.0,
+ "grad_norm": 0.00041891244472935796,
+ "learning_rate": 5.21258778926865e-05,
+ "loss": 0.0,
+ "step": 360
+ },
+ {
+ "epoch": 8.22,
+ "grad_norm": 0.0004031055432278663,
+ "learning_rate": 4.150054552753055e-05,
+ "loss": 0.0,
+ "step": 370
+ },
+ {
+ "epoch": 8.44,
+ "grad_norm": 0.00040033855475485325,
+ "learning_rate": 3.198764796404807e-05,
+ "loss": 0.0,
+ "step": 380
+ },
+ {
+ "epoch": 8.67,
+ "grad_norm": 0.0015161971095949411,
+ "learning_rate": 2.3637940626713344e-05,
+ "loss": 0.0,
+ "step": 390
+ },
+ {
+ "epoch": 8.89,
+ "grad_norm": 0.0011603726306930184,
+ "learning_rate": 1.649597281686302e-05,
+ "loss": 0.0,
+ "step": 400
+ },
+ {
+ "epoch": 8.89,
+ "eval_loss": 0.014345706440508366,
+ "eval_runtime": 4.8186,
+ "eval_samples_per_second": 39.43,
+ "eval_steps_per_second": 1.66,
+ "step": 400
+ },
+ {
+ "epoch": 9.11,
+ "grad_norm": 0.0003027151105925441,
+ "learning_rate": 1.0599850022898538e-05,
+ "loss": 0.0001,
+ "step": 410
+ },
+ {
+ "epoch": 9.33,
+ "grad_norm": 0.0009049432119354606,
+ "learning_rate": 5.981030611018234e-06,
+ "loss": 0.0,
+ "step": 420
+ },
+ {
+ "epoch": 9.56,
+ "grad_norm": 0.0018125231144949794,
+ "learning_rate": 2.664157981222437e-06,
+ "loss": 0.0,
+ "step": 430
+ },
+ {
+ "epoch": 9.78,
+ "grad_norm": 0.003249464090913534,
+ "learning_rate": 6.66929084112089e-07,
+ "loss": 0.0,
+ "step": 440
+ },
+ {
+ "epoch": 10.0,
+ "grad_norm": 0.0004782461328431964,
+ "learning_rate": 0.0,
+ "loss": 0.0,
+ "step": 450
+ },
+ {
+ "epoch": 10.0,
+ "step": 450,
+ "total_flos": 4.3483361730730394e+17,
+ "train_loss": 0.025018121934683425,
+ "train_runtime": 909.2296,
+ "train_samples_per_second": 11.834,
+ "train_steps_per_second": 0.495
+ }
+ ],
+ "logging_steps": 10,
+ "max_steps": 450,
+ "num_input_tokens_seen": 0,
+ "num_train_epochs": 10,
+ "save_steps": 500,
+ "total_flos": 4.3483361730730394e+17,
+ "train_batch_size": 8,
+ "trial_name": null,
+ "trial_params": null
+}
diff --git a/llama2_13b_peft/question_selection/training_args.bin b/llama2_13b_peft/question_selection/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..7e77de42ae41e26743f8716ef8c97a17f88ff9a0
--- /dev/null
+++ b/llama2_13b_peft/question_selection/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1534a1067a92ba562ea5440d758f3eb4b29d90267fe1882cd21507d9bbcec9e5
+size 5112
diff --git a/llama2_13b_peft/question_selection/training_eval_loss.png b/llama2_13b_peft/question_selection/training_eval_loss.png
new file mode 100644
index 0000000000000000000000000000000000000000..4088b83230376ef608570fc8a822a0e584b58b56
Binary files /dev/null and b/llama2_13b_peft/question_selection/training_eval_loss.png differ
diff --git a/llama2_13b_peft/question_selection/training_loss.png b/llama2_13b_peft/question_selection/training_loss.png
new file mode 100644
index 0000000000000000000000000000000000000000..1ba41ce12f383fb9ad7ac05baff72d0ffb137193
Binary files /dev/null and b/llama2_13b_peft/question_selection/training_loss.png differ
diff --git a/llama2_13b_peft/reasoning_about_colored_objects/README.md b/llama2_13b_peft/reasoning_about_colored_objects/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..a2914268474f9d813424a7840e3dfb2022368d58
--- /dev/null
+++ b/llama2_13b_peft/reasoning_about_colored_objects/README.md
@@ -0,0 +1,69 @@
+---
+license: other
+library_name: peft
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+base_model: /data1/model/llama2/meta-llama/Llama2-13b
+model-index:
+- name: reasoning_about_colored_objects_no_sys
+ results: []
+---
+
+
+
+# reasoning_about_colored_objects_no_sys
+
+This model is a fine-tuned version of [/data1/model/llama2/meta-llama/Llama2-13b](https://huggingface.co//data1/model/llama2/meta-llama/Llama2-13b) on the reasoning_about_colored_objects_no_sys dataset.
+It achieves the following results on the evaluation set:
+- Loss: 0.0522
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 4
+- eval_batch_size: 4
+- seed: 42
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_steps: 20
+- num_epochs: 5.0
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:------:|:----:|:---------------:|
+| 0.2126 | 0.2632 | 100 | 0.1725 |
+| 0.1793 | 0.5263 | 200 | 0.1547 |
+| 0.1496 | 0.7895 | 300 | 0.1156 |
+| 0.2432 | 1.0526 | 400 | 0.0812 |
+| 0.1287 | 1.3158 | 500 | 0.0522 |
+| 0.1829 | 1.5789 | 600 | 0.1413 |
+| 0.0235 | 1.8421 | 700 | 0.0713 |
+| 0.1101 | 2.1053 | 800 | 0.0611 |
+
+
+### Framework versions
+
+- PEFT 0.10.0
+- Transformers 4.40.0
+- Pytorch 2.2.1
+- Datasets 2.18.0
+- Tokenizers 0.19.1
\ No newline at end of file
diff --git a/llama2_13b_peft/reasoning_about_colored_objects/adapter_config.json b/llama2_13b_peft/reasoning_about_colored_objects/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..8d9114c924478ca811ae45bb660eeacec4712717
--- /dev/null
+++ b/llama2_13b_peft/reasoning_about_colored_objects/adapter_config.json
@@ -0,0 +1,34 @@
+{
+ "alpha_pattern": {},
+ "auto_mapping": null,
+ "base_model_name_or_path": "/data1/model/llama2/meta-llama/Llama2-13b",
+ "bias": "none",
+ "fan_in_fan_out": false,
+ "inference_mode": true,
+ "init_lora_weights": true,
+ "layer_replication": null,
+ "layers_pattern": null,
+ "layers_to_transform": null,
+ "loftq_config": {},
+ "lora_alpha": 16,
+ "lora_dropout": 0.0,
+ "megatron_config": null,
+ "megatron_core": "megatron.core",
+ "modules_to_save": null,
+ "peft_type": "LORA",
+ "r": 8,
+ "rank_pattern": {},
+ "revision": null,
+ "target_modules": [
+ "q_proj",
+ "gate_proj",
+ "down_proj",
+ "v_proj",
+ "up_proj",
+ "o_proj",
+ "k_proj"
+ ],
+ "task_type": "CAUSAL_LM",
+ "use_dora": false,
+ "use_rslora": false
+}
\ No newline at end of file
diff --git a/llama2_13b_peft/reasoning_about_colored_objects/adapter_model.safetensors b/llama2_13b_peft/reasoning_about_colored_objects/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..0b55315f2032f093de7a411d443862e578a4af6d
--- /dev/null
+++ b/llama2_13b_peft/reasoning_about_colored_objects/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c19495b1207cd22a199f60666c7f5a7f69fecf07adfa400406e1ca671ecae0c4
+size 125248064
diff --git a/llama2_13b_peft/reasoning_about_colored_objects/all_results.json b/llama2_13b_peft/reasoning_about_colored_objects/all_results.json
new file mode 100644
index 0000000000000000000000000000000000000000..57e29f6b1bdd76e38acde7ef8b8b51829f798fa1
--- /dev/null
+++ b/llama2_13b_peft/reasoning_about_colored_objects/all_results.json
@@ -0,0 +1,12 @@
+{
+ "epoch": 2.1052631578947367,
+ "eval_loss": 0.05217451602220535,
+ "eval_runtime": 1.5829,
+ "eval_samples_per_second": 50.54,
+ "eval_steps_per_second": 12.635,
+ "total_flos": 2.241573610389504e+16,
+ "train_loss": 0.32420646863698493,
+ "train_runtime": 246.1313,
+ "train_samples_per_second": 30.878,
+ "train_steps_per_second": 7.719
+}
\ No newline at end of file
diff --git a/llama2_13b_peft/reasoning_about_colored_objects/eval_results.json b/llama2_13b_peft/reasoning_about_colored_objects/eval_results.json
new file mode 100644
index 0000000000000000000000000000000000000000..6dbdd2c0561f3a060d471705213547f34fed819e
--- /dev/null
+++ b/llama2_13b_peft/reasoning_about_colored_objects/eval_results.json
@@ -0,0 +1,7 @@
+{
+ "epoch": 2.1052631578947367,
+ "eval_loss": 0.05217451602220535,
+ "eval_runtime": 1.5829,
+ "eval_samples_per_second": 50.54,
+ "eval_steps_per_second": 12.635
+}
\ No newline at end of file
diff --git a/llama2_13b_peft/reasoning_about_colored_objects/special_tokens_map.json b/llama2_13b_peft/reasoning_about_colored_objects/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..72ecfeeb7e14d244c936169d2ed139eeae235ef1
--- /dev/null
+++ b/llama2_13b_peft/reasoning_about_colored_objects/special_tokens_map.json
@@ -0,0 +1,24 @@
+{
+ "bos_token": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false
+ },
+ "eos_token": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false
+ },
+ "pad_token": "",
+ "unk_token": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false
+ }
+}
diff --git a/llama2_13b_peft/reasoning_about_colored_objects/tokenizer.model b/llama2_13b_peft/reasoning_about_colored_objects/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..6c00c742ce03c627d6cd5b795984876fa49fa899
--- /dev/null
+++ b/llama2_13b_peft/reasoning_about_colored_objects/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723
diff --git a/llama2_13b_peft/reasoning_about_colored_objects/tokenizer_config.json b/llama2_13b_peft/reasoning_about_colored_objects/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..a40266f39e5b5fed14de34710d35eb9e98d6bdad
--- /dev/null
+++ b/llama2_13b_peft/reasoning_about_colored_objects/tokenizer_config.json
@@ -0,0 +1,45 @@
+{
+ "add_bos_token": true,
+ "add_eos_token": false,
+ "add_prefix_space": true,
+ "added_tokens_decoder": {
+ "0": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false,
+ "special": true
+ },
+ "1": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false,
+ "special": true
+ },
+ "2": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false,
+ "special": true
+ }
+ },
+ "bos_token": "",
+ "chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '\\n' }}{% endif %}{% endfor %}",
+ "clean_up_tokenization_spaces": false,
+ "eos_token": "",
+ "legacy": true,
+ "model_max_length": 1000000000000000019884624838656,
+ "pad_token": "",
+ "padding_side": "right",
+ "sp_model_kwargs": {},
+ "spaces_between_special_tokens": false,
+ "split_special_tokens": false,
+ "tokenizer_class": "LlamaTokenizer",
+ "unk_token": "",
+ "use_default_system_prompt": false
+}
diff --git a/llama2_13b_peft/reasoning_about_colored_objects/train_results.json b/llama2_13b_peft/reasoning_about_colored_objects/train_results.json
new file mode 100644
index 0000000000000000000000000000000000000000..033f03cbfd9d9032887c66ef117290197e915470
--- /dev/null
+++ b/llama2_13b_peft/reasoning_about_colored_objects/train_results.json
@@ -0,0 +1,8 @@
+{
+ "epoch": 2.1052631578947367,
+ "total_flos": 2.241573610389504e+16,
+ "train_loss": 0.32420646863698493,
+ "train_runtime": 246.1313,
+ "train_samples_per_second": 30.878,
+ "train_steps_per_second": 7.719
+}
\ No newline at end of file
diff --git a/llama2_13b_peft/reasoning_about_colored_objects/trainer_log.jsonl b/llama2_13b_peft/reasoning_about_colored_objects/trainer_log.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..5c70c0848a19baed24a440553a499f77edd45d99
--- /dev/null
+++ b/llama2_13b_peft/reasoning_about_colored_objects/trainer_log.jsonl
@@ -0,0 +1,90 @@
+{"current_steps": 10, "total_steps": 1900, "loss": 7.977, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 2.5e-05, "epoch": 0.02631578947368421, "percentage": 0.53, "elapsed_time": "0:00:04", "remaining_time": "0:13:00"}
+{"current_steps": 20, "total_steps": 1900, "loss": 6.4447, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 5e-05, "epoch": 0.05263157894736842, "percentage": 1.05, "elapsed_time": "0:00:06", "remaining_time": "0:10:31"}
+{"current_steps": 30, "total_steps": 1900, "loss": 1.6555, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.999650952964643e-05, "epoch": 0.07894736842105263, "percentage": 1.58, "elapsed_time": "0:00:09", "remaining_time": "0:09:41"}
+{"current_steps": 40, "total_steps": 1900, "loss": 0.4157, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.998603909325637e-05, "epoch": 0.10526315789473684, "percentage": 2.11, "elapsed_time": "0:00:11", "remaining_time": "0:09:14"}
+{"current_steps": 50, "total_steps": 1900, "loss": 0.4377, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.996859161456965e-05, "epoch": 0.13157894736842105, "percentage": 2.63, "elapsed_time": "0:00:14", "remaining_time": "0:08:57"}
+{"current_steps": 60, "total_steps": 1900, "loss": 0.4515, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.9944171965578836e-05, "epoch": 0.15789473684210525, "percentage": 3.16, "elapsed_time": "0:00:17", "remaining_time": "0:08:46"}
+{"current_steps": 70, "total_steps": 1900, "loss": 0.2769, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.991278696516879e-05, "epoch": 0.18421052631578946, "percentage": 3.68, "elapsed_time": "0:00:19", "remaining_time": "0:08:37"}
+{"current_steps": 80, "total_steps": 1900, "loss": 0.3058, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.9874445377212606e-05, "epoch": 0.21052631578947367, "percentage": 4.21, "elapsed_time": "0:00:22", "remaining_time": "0:08:31"}
+{"current_steps": 90, "total_steps": 1900, "loss": 0.3412, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.9829157908124355e-05, "epoch": 0.23684210526315788, "percentage": 4.74, "elapsed_time": "0:00:25", "remaining_time": "0:08:24"}
+{"current_steps": 100, "total_steps": 1900, "loss": 0.2126, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.97769372038695e-05, "epoch": 0.2631578947368421, "percentage": 5.26, "elapsed_time": "0:00:27", "remaining_time": "0:08:19"}
+{"current_steps": 100, "total_steps": 1900, "loss": null, "eval_loss": 0.1724734604358673, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 0.2631578947368421, "percentage": 5.26, "elapsed_time": "0:00:27", "remaining_time": "0:08:19"}
+{"current_steps": 110, "total_steps": 1900, "loss": 0.1496, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.971779784643365e-05, "epoch": 0.2894736842105263, "percentage": 5.79, "elapsed_time": "0:00:32", "remaining_time": "0:08:47"}
+{"current_steps": 120, "total_steps": 1900, "loss": 0.1157, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.9651756349750716e-05, "epoch": 0.3157894736842105, "percentage": 6.32, "elapsed_time": "0:00:35", "remaining_time": "0:08:40"}
+{"current_steps": 130, "total_steps": 1900, "loss": 0.1293, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.957883115509159e-05, "epoch": 0.34210526315789475, "percentage": 6.84, "elapsed_time": "0:00:37", "remaining_time": "0:08:33"}
+{"current_steps": 140, "total_steps": 1900, "loss": 0.1512, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.9499042625914674e-05, "epoch": 0.3684210526315789, "percentage": 7.37, "elapsed_time": "0:00:40", "remaining_time": "0:08:27"}
+{"current_steps": 150, "total_steps": 1900, "loss": 0.2275, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.9412413042179615e-05, "epoch": 0.39473684210526316, "percentage": 7.89, "elapsed_time": "0:00:43", "remaining_time": "0:08:22"}
+{"current_steps": 160, "total_steps": 1900, "loss": 0.1408, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.931896659412593e-05, "epoch": 0.42105263157894735, "percentage": 8.42, "elapsed_time": "0:00:45", "remaining_time": "0:08:17"}
+{"current_steps": 170, "total_steps": 1900, "loss": 0.1723, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.9218729375518135e-05, "epoch": 0.4473684210526316, "percentage": 8.95, "elapsed_time": "0:00:48", "remaining_time": "0:08:12"}
+{"current_steps": 180, "total_steps": 1900, "loss": 0.0541, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.911172937635942e-05, "epoch": 0.47368421052631576, "percentage": 9.47, "elapsed_time": "0:00:51", "remaining_time": "0:08:07"}
+{"current_steps": 190, "total_steps": 1900, "loss": 0.1985, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.899799647507577e-05, "epoch": 0.5, "percentage": 10.0, "elapsed_time": "0:00:53", "remaining_time": "0:08:03"}
+{"current_steps": 200, "total_steps": 1900, "loss": 0.1793, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.8877562430172815e-05, "epoch": 0.5263157894736842, "percentage": 10.53, "elapsed_time": "0:00:56", "remaining_time": "0:07:59"}
+{"current_steps": 200, "total_steps": 1900, "loss": null, "eval_loss": 0.15467724204063416, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 0.5263157894736842, "percentage": 10.53, "elapsed_time": "0:00:56", "remaining_time": "0:07:59"}
+{"current_steps": 210, "total_steps": 1900, "loss": 0.1591, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.8750460871367605e-05, "epoch": 0.5526315789473685, "percentage": 11.05, "elapsed_time": "0:01:01", "remaining_time": "0:08:11"}
+{"current_steps": 220, "total_steps": 1900, "loss": 0.1407, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.861672729019797e-05, "epoch": 0.5789473684210527, "percentage": 11.58, "elapsed_time": "0:01:03", "remaining_time": "0:08:06"}
+{"current_steps": 230, "total_steps": 1900, "loss": 0.1257, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.847639903011195e-05, "epoch": 0.6052631578947368, "percentage": 12.11, "elapsed_time": "0:01:06", "remaining_time": "0:08:02"}
+{"current_steps": 240, "total_steps": 1900, "loss": 0.1254, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.832951527604007e-05, "epoch": 0.631578947368421, "percentage": 12.63, "elapsed_time": "0:01:09", "remaining_time": "0:07:58"}
+{"current_steps": 250, "total_steps": 1900, "loss": 0.1557, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.817611704345344e-05, "epoch": 0.6578947368421053, "percentage": 13.16, "elapsed_time": "0:01:11", "remaining_time": "0:07:53"}
+{"current_steps": 260, "total_steps": 1900, "loss": 0.1344, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.801624716691072e-05, "epoch": 0.6842105263157895, "percentage": 13.68, "elapsed_time": "0:01:14", "remaining_time": "0:07:49"}
+{"current_steps": 270, "total_steps": 1900, "loss": 0.4058, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.784995028809707e-05, "epoch": 0.7105263157894737, "percentage": 14.21, "elapsed_time": "0:01:17", "remaining_time": "0:07:45"}
+{"current_steps": 280, "total_steps": 1900, "loss": 0.0956, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.767727284335852e-05, "epoch": 0.7368421052631579, "percentage": 14.74, "elapsed_time": "0:01:19", "remaining_time": "0:07:41"}
+{"current_steps": 290, "total_steps": 1900, "loss": 0.1058, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.74982630507352e-05, "epoch": 0.7631578947368421, "percentage": 15.26, "elapsed_time": "0:01:22", "remaining_time": "0:07:37"}
+{"current_steps": 300, "total_steps": 1900, "loss": 0.1496, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.731297089649703e-05, "epoch": 0.7894736842105263, "percentage": 15.79, "elapsed_time": "0:01:24", "remaining_time": "0:07:33"}
+{"current_steps": 300, "total_steps": 1900, "loss": null, "eval_loss": 0.11555701494216919, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 0.7894736842105263, "percentage": 15.79, "elapsed_time": "0:01:24", "remaining_time": "0:07:33"}
+{"current_steps": 310, "total_steps": 1900, "loss": 0.1453, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.712144812118572e-05, "epoch": 0.8157894736842105, "percentage": 16.32, "elapsed_time": "0:01:29", "remaining_time": "0:07:39"}
+{"current_steps": 320, "total_steps": 1900, "loss": 0.0982, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.692374820516679e-05, "epoch": 0.8421052631578947, "percentage": 16.84, "elapsed_time": "0:01:32", "remaining_time": "0:07:35"}
+{"current_steps": 330, "total_steps": 1900, "loss": 0.1817, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.671992635369591e-05, "epoch": 0.868421052631579, "percentage": 17.37, "elapsed_time": "0:01:34", "remaining_time": "0:07:31"}
+{"current_steps": 340, "total_steps": 1900, "loss": 0.0288, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.651003948150349e-05, "epoch": 0.8947368421052632, "percentage": 17.89, "elapsed_time": "0:01:37", "remaining_time": "0:07:27"}
+{"current_steps": 350, "total_steps": 1900, "loss": 0.3321, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.62941461969019e-05, "epoch": 0.9210526315789473, "percentage": 18.42, "elapsed_time": "0:01:40", "remaining_time": "0:07:23"}
+{"current_steps": 360, "total_steps": 1900, "loss": 0.1525, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.6072306785419926e-05, "epoch": 0.9473684210526315, "percentage": 18.95, "elapsed_time": "0:01:42", "remaining_time": "0:07:19"}
+{"current_steps": 370, "total_steps": 1900, "loss": 0.0626, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.584458319296868e-05, "epoch": 0.9736842105263158, "percentage": 19.47, "elapsed_time": "0:01:45", "remaining_time": "0:07:15"}
+{"current_steps": 380, "total_steps": 1900, "loss": 0.0961, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.561103900854401e-05, "epoch": 1.0, "percentage": 20.0, "elapsed_time": "0:01:48", "remaining_time": "0:07:12"}
+{"current_steps": 390, "total_steps": 1900, "loss": 0.092, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.537173944647008e-05, "epoch": 1.0263157894736843, "percentage": 20.53, "elapsed_time": "0:01:50", "remaining_time": "0:07:08"}
+{"current_steps": 400, "total_steps": 1900, "loss": 0.2432, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.512675132818908e-05, "epoch": 1.0526315789473684, "percentage": 21.05, "elapsed_time": "0:01:53", "remaining_time": "0:07:04"}
+{"current_steps": 400, "total_steps": 1900, "loss": null, "eval_loss": 0.08122994005680084, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 1.0526315789473684, "percentage": 21.05, "elapsed_time": "0:01:53", "remaining_time": "0:07:04"}
+{"current_steps": 410, "total_steps": 1900, "loss": 0.0205, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.487614306360208e-05, "epoch": 1.0789473684210527, "percentage": 21.58, "elapsed_time": "0:01:57", "remaining_time": "0:07:08"}
+{"current_steps": 420, "total_steps": 1900, "loss": 0.079, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.4619984631966524e-05, "epoch": 1.1052631578947367, "percentage": 22.11, "elapsed_time": "0:02:00", "remaining_time": "0:07:04"}
+{"current_steps": 430, "total_steps": 1900, "loss": 0.0662, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.435834756235534e-05, "epoch": 1.131578947368421, "percentage": 22.63, "elapsed_time": "0:02:03", "remaining_time": "0:07:00"}
+{"current_steps": 440, "total_steps": 1900, "loss": 0.1356, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.409130491368331e-05, "epoch": 1.1578947368421053, "percentage": 23.16, "elapsed_time": "0:02:05", "remaining_time": "0:06:57"}
+{"current_steps": 450, "total_steps": 1900, "loss": 0.1175, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.381893125430628e-05, "epoch": 1.1842105263157894, "percentage": 23.68, "elapsed_time": "0:02:08", "remaining_time": "0:06:53"}
+{"current_steps": 460, "total_steps": 1900, "loss": 0.2242, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.354130264119894e-05, "epoch": 1.2105263157894737, "percentage": 24.21, "elapsed_time": "0:02:10", "remaining_time": "0:06:50"}
+{"current_steps": 470, "total_steps": 1900, "loss": 0.0083, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.3258496598716736e-05, "epoch": 1.236842105263158, "percentage": 24.74, "elapsed_time": "0:02:13", "remaining_time": "0:06:46"}
+{"current_steps": 480, "total_steps": 1900, "loss": 0.0576, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.2970592096948236e-05, "epoch": 1.263157894736842, "percentage": 25.26, "elapsed_time": "0:02:16", "remaining_time": "0:06:43"}
+{"current_steps": 490, "total_steps": 1900, "loss": 0.0405, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.267766952966369e-05, "epoch": 1.2894736842105263, "percentage": 25.79, "elapsed_time": "0:02:18", "remaining_time": "0:06:39"}
+{"current_steps": 500, "total_steps": 1900, "loss": 0.1287, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.2379810691866064e-05, "epoch": 1.3157894736842106, "percentage": 26.32, "elapsed_time": "0:02:21", "remaining_time": "0:06:36"}
+{"current_steps": 500, "total_steps": 1900, "loss": null, "eval_loss": 0.05217451602220535, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 1.3157894736842106, "percentage": 26.32, "elapsed_time": "0:02:21", "remaining_time": "0:06:36"}
+{"current_steps": 510, "total_steps": 1900, "loss": 0.0409, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.2077098756950775e-05, "epoch": 1.3421052631578947, "percentage": 26.84, "elapsed_time": "0:02:26", "remaining_time": "0:06:38"}
+{"current_steps": 520, "total_steps": 1900, "loss": 0.0092, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.176961825348059e-05, "epoch": 1.368421052631579, "percentage": 27.37, "elapsed_time": "0:02:28", "remaining_time": "0:06:34"}
+{"current_steps": 530, "total_steps": 1900, "loss": 0.0337, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.1457455041582044e-05, "epoch": 1.3947368421052633, "percentage": 27.89, "elapsed_time": "0:02:31", "remaining_time": "0:06:31"}
+{"current_steps": 540, "total_steps": 1900, "loss": 0.0519, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.114069628897006e-05, "epoch": 1.4210526315789473, "percentage": 28.42, "elapsed_time": "0:02:33", "remaining_time": "0:06:27"}
+{"current_steps": 550, "total_steps": 1900, "loss": 0.0814, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.0819430446607465e-05, "epoch": 1.4473684210526316, "percentage": 28.95, "elapsed_time": "0:02:36", "remaining_time": "0:06:24"}
+{"current_steps": 560, "total_steps": 1900, "loss": 0.0564, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.049374722400612e-05, "epoch": 1.4736842105263157, "percentage": 29.47, "elapsed_time": "0:02:39", "remaining_time": "0:06:21"}
+{"current_steps": 570, "total_steps": 1900, "loss": 0.0943, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 4.016373756417669e-05, "epoch": 1.5, "percentage": 30.0, "elapsed_time": "0:02:42", "remaining_time": "0:06:18"}
+{"current_steps": 580, "total_steps": 1900, "loss": 0.0423, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.982949361823388e-05, "epoch": 1.526315789473684, "percentage": 30.53, "elapsed_time": "0:02:44", "remaining_time": "0:06:14"}
+{"current_steps": 590, "total_steps": 1900, "loss": 0.0005, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.949110871966444e-05, "epoch": 1.5526315789473686, "percentage": 31.05, "elapsed_time": "0:02:47", "remaining_time": "0:06:11"}
+{"current_steps": 600, "total_steps": 1900, "loss": 0.1829, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.914867735826489e-05, "epoch": 1.5789473684210527, "percentage": 31.58, "elapsed_time": "0:02:49", "remaining_time": "0:06:08"}
+{"current_steps": 600, "total_steps": 1900, "loss": null, "eval_loss": 0.1412845253944397, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 1.5789473684210527, "percentage": 31.58, "elapsed_time": "0:02:49", "remaining_time": "0:06:08"}
+{"current_steps": 610, "total_steps": 1900, "loss": 0.0851, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.880229515375642e-05, "epoch": 1.6052631578947367, "percentage": 32.11, "elapsed_time": "0:02:54", "remaining_time": "0:06:09"}
+{"current_steps": 620, "total_steps": 1900, "loss": 0.156, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.845205882908432e-05, "epoch": 1.631578947368421, "percentage": 32.63, "elapsed_time": "0:02:57", "remaining_time": "0:06:05"}
+{"current_steps": 630, "total_steps": 1900, "loss": 0.0431, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.809806618340922e-05, "epoch": 1.6578947368421053, "percentage": 33.16, "elapsed_time": "0:02:59", "remaining_time": "0:06:02"}
+{"current_steps": 640, "total_steps": 1900, "loss": 0.057, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.774041606479794e-05, "epoch": 1.6842105263157894, "percentage": 33.68, "elapsed_time": "0:03:02", "remaining_time": "0:05:59"}
+{"current_steps": 650, "total_steps": 1900, "loss": 0.0936, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.737920834262134e-05, "epoch": 1.7105263157894737, "percentage": 34.21, "elapsed_time": "0:03:05", "remaining_time": "0:05:55"}
+{"current_steps": 660, "total_steps": 1900, "loss": 0.0532, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.7014543879667094e-05, "epoch": 1.736842105263158, "percentage": 34.74, "elapsed_time": "0:03:07", "remaining_time": "0:05:52"}
+{"current_steps": 670, "total_steps": 1900, "loss": 0.0547, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.6646524503974955e-05, "epoch": 1.763157894736842, "percentage": 35.26, "elapsed_time": "0:03:10", "remaining_time": "0:05:49"}
+{"current_steps": 680, "total_steps": 1900, "loss": 0.1166, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.6275252980402544e-05, "epoch": 1.7894736842105263, "percentage": 35.79, "elapsed_time": "0:03:12", "remaining_time": "0:05:46"}
+{"current_steps": 690, "total_steps": 1900, "loss": 0.0327, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.590083298192957e-05, "epoch": 1.8157894736842106, "percentage": 36.32, "elapsed_time": "0:03:15", "remaining_time": "0:05:42"}
+{"current_steps": 700, "total_steps": 1900, "loss": 0.0235, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.552336906070838e-05, "epoch": 1.8421052631578947, "percentage": 36.84, "elapsed_time": "0:03:18", "remaining_time": "0:05:39"}
+{"current_steps": 700, "total_steps": 1900, "loss": null, "eval_loss": 0.07125674188137054, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 1.8421052631578947, "percentage": 36.84, "elapsed_time": "0:03:18", "remaining_time": "0:05:39"}
+{"current_steps": 710, "total_steps": 1900, "loss": 0.0007, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.51429666188691e-05, "epoch": 1.868421052631579, "percentage": 37.37, "elapsed_time": "0:03:22", "remaining_time": "0:05:39"}
+{"current_steps": 720, "total_steps": 1900, "loss": 0.1828, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.475973187908737e-05, "epoch": 1.8947368421052633, "percentage": 37.89, "elapsed_time": "0:03:25", "remaining_time": "0:05:36"}
+{"current_steps": 730, "total_steps": 1900, "loss": 0.0011, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.437377185492303e-05, "epoch": 1.9210526315789473, "percentage": 38.42, "elapsed_time": "0:03:28", "remaining_time": "0:05:33"}
+{"current_steps": 740, "total_steps": 1900, "loss": 0.1991, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.398519432093782e-05, "epoch": 1.9473684210526314, "percentage": 38.95, "elapsed_time": "0:03:30", "remaining_time": "0:05:30"}
+{"current_steps": 750, "total_steps": 1900, "loss": 0.0353, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.3594107782600753e-05, "epoch": 1.973684210526316, "percentage": 39.47, "elapsed_time": "0:03:33", "remaining_time": "0:05:27"}
+{"current_steps": 760, "total_steps": 1900, "loss": 0.0441, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.3200621445989226e-05, "epoch": 2.0, "percentage": 40.0, "elapsed_time": "0:03:35", "remaining_time": "0:05:23"}
+{"current_steps": 770, "total_steps": 1900, "loss": 0.0672, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.280484518729466e-05, "epoch": 2.026315789473684, "percentage": 40.53, "elapsed_time": "0:03:38", "remaining_time": "0:05:20"}
+{"current_steps": 780, "total_steps": 1900, "loss": 0.0298, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.2406889522140856e-05, "epoch": 2.0526315789473686, "percentage": 41.05, "elapsed_time": "0:03:41", "remaining_time": "0:05:17"}
+{"current_steps": 790, "total_steps": 1900, "loss": 0.0121, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.200686557472391e-05, "epoch": 2.0789473684210527, "percentage": 41.58, "elapsed_time": "0:03:43", "remaining_time": "0:05:14"}
+{"current_steps": 800, "total_steps": 1900, "loss": 0.1101, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": 3.160488504678216e-05, "epoch": 2.1052631578947367, "percentage": 42.11, "elapsed_time": "0:03:46", "remaining_time": "0:05:11"}
+{"current_steps": 800, "total_steps": 1900, "loss": null, "eval_loss": 0.06111542135477066, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 2.1052631578947367, "percentage": 42.11, "elapsed_time": "0:03:46", "remaining_time": "0:05:11"}
+{"current_steps": 800, "total_steps": 1900, "loss": null, "eval_loss": null, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 2.1052631578947367, "percentage": 42.11, "elapsed_time": "0:03:46", "remaining_time": "0:05:11"}
+{"current_steps": 20, "total_steps": 20, "loss": null, "eval_loss": 0.05217451602220535, "predict_loss": null, "reward": null, "accuracy": null, "learning_rate": null, "epoch": 2.1052631578947367, "percentage": 100.0, "elapsed_time": "0:03:50", "remaining_time": "0:00:00"}
diff --git a/llama2_13b_peft/reasoning_about_colored_objects/trainer_state.json b/llama2_13b_peft/reasoning_about_colored_objects/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..15506eccc5b3fd9cd3bf1041d3d172c089a3f9b2
--- /dev/null
+++ b/llama2_13b_peft/reasoning_about_colored_objects/trainer_state.json
@@ -0,0 +1,654 @@
+{
+ "best_metric": 0.05217451602220535,
+ "best_model_checkpoint": "ckpt/llama2_13b_fuze27_no_sys/reasoning_about_colored_objects_no_sys/checkpoint-500",
+ "epoch": 2.1052631578947367,
+ "eval_steps": 100,
+ "global_step": 800,
+ "is_hyper_param_search": false,
+ "is_local_process_zero": true,
+ "is_world_process_zero": true,
+ "log_history": [
+ {
+ "epoch": 0.02631578947368421,
+ "grad_norm": 5.66915225982666,
+ "learning_rate": 2.5e-05,
+ "loss": 7.977,
+ "step": 10
+ },
+ {
+ "epoch": 0.05263157894736842,
+ "grad_norm": 6.004359245300293,
+ "learning_rate": 5e-05,
+ "loss": 6.4447,
+ "step": 20
+ },
+ {
+ "epoch": 0.07894736842105263,
+ "grad_norm": 3.6544461250305176,
+ "learning_rate": 4.999650952964643e-05,
+ "loss": 1.6555,
+ "step": 30
+ },
+ {
+ "epoch": 0.10526315789473684,
+ "grad_norm": 5.159526348114014,
+ "learning_rate": 4.998603909325637e-05,
+ "loss": 0.4157,
+ "step": 40
+ },
+ {
+ "epoch": 0.13157894736842105,
+ "grad_norm": 4.8554792404174805,
+ "learning_rate": 4.996859161456965e-05,
+ "loss": 0.4377,
+ "step": 50
+ },
+ {
+ "epoch": 0.15789473684210525,
+ "grad_norm": 1.000138759613037,
+ "learning_rate": 4.9944171965578836e-05,
+ "loss": 0.4515,
+ "step": 60
+ },
+ {
+ "epoch": 0.18421052631578946,
+ "grad_norm": 4.686236381530762,
+ "learning_rate": 4.991278696516879e-05,
+ "loss": 0.2769,
+ "step": 70
+ },
+ {
+ "epoch": 0.21052631578947367,
+ "grad_norm": 3.694870948791504,
+ "learning_rate": 4.9874445377212606e-05,
+ "loss": 0.3058,
+ "step": 80
+ },
+ {
+ "epoch": 0.23684210526315788,
+ "grad_norm": 3.9635472297668457,
+ "learning_rate": 4.9829157908124355e-05,
+ "loss": 0.3412,
+ "step": 90
+ },
+ {
+ "epoch": 0.2631578947368421,
+ "grad_norm": 4.468523979187012,
+ "learning_rate": 4.97769372038695e-05,
+ "loss": 0.2126,
+ "step": 100
+ },
+ {
+ "epoch": 0.2631578947368421,
+ "eval_loss": 0.1724734604358673,
+ "eval_runtime": 1.5853,
+ "eval_samples_per_second": 50.465,
+ "eval_steps_per_second": 12.616,
+ "step": 100
+ },
+ {
+ "epoch": 0.2894736842105263,
+ "grad_norm": 0.02684151381254196,
+ "learning_rate": 4.971779784643365e-05,
+ "loss": 0.1496,
+ "step": 110
+ },
+ {
+ "epoch": 0.3157894736842105,
+ "grad_norm": 0.4909035265445709,
+ "learning_rate": 4.9651756349750716e-05,
+ "loss": 0.1157,
+ "step": 120
+ },
+ {
+ "epoch": 0.34210526315789475,
+ "grad_norm": 0.3278777599334717,
+ "learning_rate": 4.957883115509159e-05,
+ "loss": 0.1293,
+ "step": 130
+ },
+ {
+ "epoch": 0.3684210526315789,
+ "grad_norm": 0.08337876945734024,
+ "learning_rate": 4.9499042625914674e-05,
+ "loss": 0.1512,
+ "step": 140
+ },
+ {
+ "epoch": 0.39473684210526316,
+ "grad_norm": 2.6098978519439697,
+ "learning_rate": 4.9412413042179615e-05,
+ "loss": 0.2275,
+ "step": 150
+ },
+ {
+ "epoch": 0.42105263157894735,
+ "grad_norm": 0.006517685949802399,
+ "learning_rate": 4.931896659412593e-05,
+ "loss": 0.1408,
+ "step": 160
+ },
+ {
+ "epoch": 0.4473684210526316,
+ "grad_norm": 3.060093641281128,
+ "learning_rate": 4.9218729375518135e-05,
+ "loss": 0.1723,
+ "step": 170
+ },
+ {
+ "epoch": 0.47368421052631576,
+ "grad_norm": 1.4745004177093506,
+ "learning_rate": 4.911172937635942e-05,
+ "loss": 0.0541,
+ "step": 180
+ },
+ {
+ "epoch": 0.5,
+ "grad_norm": 5.701627254486084,
+ "learning_rate": 4.899799647507577e-05,
+ "loss": 0.1985,
+ "step": 190
+ },
+ {
+ "epoch": 0.5263157894736842,
+ "grad_norm": 3.7217190265655518,
+ "learning_rate": 4.8877562430172815e-05,
+ "loss": 0.1793,
+ "step": 200
+ },
+ {
+ "epoch": 0.5263157894736842,
+ "eval_loss": 0.15467724204063416,
+ "eval_runtime": 1.5815,
+ "eval_samples_per_second": 50.586,
+ "eval_steps_per_second": 12.646,
+ "step": 200
+ },
+ {
+ "epoch": 0.5526315789473685,
+ "grad_norm": 0.00879894383251667,
+ "learning_rate": 4.8750460871367605e-05,
+ "loss": 0.1591,
+ "step": 210
+ },
+ {
+ "epoch": 0.5789473684210527,
+ "grad_norm": 2.3962349891662598,
+ "learning_rate": 4.861672729019797e-05,
+ "loss": 0.1407,
+ "step": 220
+ },
+ {
+ "epoch": 0.6052631578947368,
+ "grad_norm": 4.49237585067749,
+ "learning_rate": 4.847639903011195e-05,
+ "loss": 0.1257,
+ "step": 230
+ },
+ {
+ "epoch": 0.631578947368421,
+ "grad_norm": 0.023957636207342148,
+ "learning_rate": 4.832951527604007e-05,
+ "loss": 0.1254,
+ "step": 240
+ },
+ {
+ "epoch": 0.6578947368421053,
+ "grad_norm": 0.7816401124000549,
+ "learning_rate": 4.817611704345344e-05,
+ "loss": 0.1557,
+ "step": 250
+ },
+ {
+ "epoch": 0.6842105263157895,
+ "grad_norm": 5.848673343658447,
+ "learning_rate": 4.801624716691072e-05,
+ "loss": 0.1344,
+ "step": 260
+ },
+ {
+ "epoch": 0.7105263157894737,
+ "grad_norm": 9.321337699890137,
+ "learning_rate": 4.784995028809707e-05,
+ "loss": 0.4058,
+ "step": 270
+ },
+ {
+ "epoch": 0.7368421052631579,
+ "grad_norm": 5.277842998504639,
+ "learning_rate": 4.767727284335852e-05,
+ "loss": 0.0956,
+ "step": 280
+ },
+ {
+ "epoch": 0.7631578947368421,
+ "grad_norm": 10.428375244140625,
+ "learning_rate": 4.74982630507352e-05,
+ "loss": 0.1058,
+ "step": 290
+ },
+ {
+ "epoch": 0.7894736842105263,
+ "grad_norm": 3.061192750930786,
+ "learning_rate": 4.731297089649703e-05,
+ "loss": 0.1496,
+ "step": 300
+ },
+ {
+ "epoch": 0.7894736842105263,
+ "eval_loss": 0.11555701494216919,
+ "eval_runtime": 1.5784,
+ "eval_samples_per_second": 50.683,
+ "eval_steps_per_second": 12.671,
+ "step": 300
+ },
+ {
+ "epoch": 0.8157894736842105,
+ "grad_norm": 2.538578748703003,
+ "learning_rate": 4.712144812118572e-05,
+ "loss": 0.1453,
+ "step": 310
+ },
+ {
+ "epoch": 0.8421052631578947,
+ "grad_norm": 1.092056155204773,
+ "learning_rate": 4.692374820516679e-05,
+ "loss": 0.0982,
+ "step": 320
+ },
+ {
+ "epoch": 0.868421052631579,
+ "grad_norm": 8.68082332611084,
+ "learning_rate": 4.671992635369591e-05,
+ "loss": 0.1817,
+ "step": 330
+ },
+ {
+ "epoch": 0.8947368421052632,
+ "grad_norm": 0.0909847840666771,
+ "learning_rate": 4.651003948150349e-05,
+ "loss": 0.0288,
+ "step": 340
+ },
+ {
+ "epoch": 0.9210526315789473,
+ "grad_norm": 0.002436779672279954,
+ "learning_rate": 4.62941461969019e-05,
+ "loss": 0.3321,
+ "step": 350
+ },
+ {
+ "epoch": 0.9473684210526315,
+ "grad_norm": 0.018777817487716675,
+ "learning_rate": 4.6072306785419926e-05,
+ "loss": 0.1525,
+ "step": 360
+ },
+ {
+ "epoch": 0.9736842105263158,
+ "grad_norm": 4.674654006958008,
+ "learning_rate": 4.584458319296868e-05,
+ "loss": 0.0626,
+ "step": 370
+ },
+ {
+ "epoch": 1.0,
+ "grad_norm": 0.008717444725334644,
+ "learning_rate": 4.561103900854401e-05,
+ "loss": 0.0961,
+ "step": 380
+ },
+ {
+ "epoch": 1.0263157894736843,
+ "grad_norm": 5.005481243133545,
+ "learning_rate": 4.537173944647008e-05,
+ "loss": 0.092,
+ "step": 390
+ },
+ {
+ "epoch": 1.0526315789473684,
+ "grad_norm": 1.9978797435760498,
+ "learning_rate": 4.512675132818908e-05,
+ "loss": 0.2432,
+ "step": 400
+ },
+ {
+ "epoch": 1.0526315789473684,
+ "eval_loss": 0.08122994005680084,
+ "eval_runtime": 1.582,
+ "eval_samples_per_second": 50.569,
+ "eval_steps_per_second": 12.642,
+ "step": 400
+ },
+ {
+ "epoch": 1.0789473684210527,
+ "grad_norm": 0.004451519809663296,
+ "learning_rate": 4.487614306360208e-05,
+ "loss": 0.0205,
+ "step": 410
+ },
+ {
+ "epoch": 1.1052631578947367,
+ "grad_norm": 3.3186466693878174,
+ "learning_rate": 4.4619984631966524e-05,
+ "loss": 0.079,
+ "step": 420
+ },
+ {
+ "epoch": 1.131578947368421,
+ "grad_norm": 0.07870589941740036,
+ "learning_rate": 4.435834756235534e-05,
+ "loss": 0.0662,
+ "step": 430
+ },
+ {
+ "epoch": 1.1578947368421053,
+ "grad_norm": 0.0030028028413653374,
+ "learning_rate": 4.409130491368331e-05,
+ "loss": 0.1356,
+ "step": 440
+ },
+ {
+ "epoch": 1.1842105263157894,
+ "grad_norm": 0.9092045426368713,
+ "learning_rate": 4.381893125430628e-05,
+ "loss": 0.1175,
+ "step": 450
+ },
+ {
+ "epoch": 1.2105263157894737,
+ "grad_norm": 10.507440567016602,
+ "learning_rate": 4.354130264119894e-05,
+ "loss": 0.2242,
+ "step": 460
+ },
+ {
+ "epoch": 1.236842105263158,
+ "grad_norm": 0.006735638249665499,
+ "learning_rate": 4.3258496598716736e-05,
+ "loss": 0.0083,
+ "step": 470
+ },
+ {
+ "epoch": 1.263157894736842,
+ "grad_norm": 4.7078471183776855,
+ "learning_rate": 4.2970592096948236e-05,
+ "loss": 0.0576,
+ "step": 480
+ },
+ {
+ "epoch": 1.2894736842105263,
+ "grad_norm": 0.0016024401411414146,
+ "learning_rate": 4.267766952966369e-05,
+ "loss": 0.0405,
+ "step": 490
+ },
+ {
+ "epoch": 1.3157894736842106,
+ "grad_norm": 0.002612845739349723,
+ "learning_rate": 4.2379810691866064e-05,
+ "loss": 0.1287,
+ "step": 500
+ },
+ {
+ "epoch": 1.3157894736842106,
+ "eval_loss": 0.05217451602220535,
+ "eval_runtime": 1.5811,
+ "eval_samples_per_second": 50.599,
+ "eval_steps_per_second": 12.65,
+ "step": 500
+ },
+ {
+ "epoch": 1.3421052631578947,
+ "grad_norm": 0.9947099089622498,
+ "learning_rate": 4.2077098756950775e-05,
+ "loss": 0.0409,
+ "step": 510
+ },
+ {
+ "epoch": 1.368421052631579,
+ "grad_norm": 0.0010769988875836134,
+ "learning_rate": 4.176961825348059e-05,
+ "loss": 0.0092,
+ "step": 520
+ },
+ {
+ "epoch": 1.3947368421052633,
+ "grad_norm": 0.03632188215851784,
+ "learning_rate": 4.1457455041582044e-05,
+ "loss": 0.0337,
+ "step": 530
+ },
+ {
+ "epoch": 1.4210526315789473,
+ "grad_norm": 0.07263629138469696,
+ "learning_rate": 4.114069628897006e-05,
+ "loss": 0.0519,
+ "step": 540
+ },
+ {
+ "epoch": 1.4473684210526316,
+ "grad_norm": 11.557324409484863,
+ "learning_rate": 4.0819430446607465e-05,
+ "loss": 0.0814,
+ "step": 550
+ },
+ {
+ "epoch": 1.4736842105263157,
+ "grad_norm": 0.0008104212465696037,
+ "learning_rate": 4.049374722400612e-05,
+ "loss": 0.0564,
+ "step": 560
+ },
+ {
+ "epoch": 1.5,
+ "grad_norm": 27.804492950439453,
+ "learning_rate": 4.016373756417669e-05,
+ "loss": 0.0943,
+ "step": 570
+ },
+ {
+ "epoch": 1.526315789473684,
+ "grad_norm": 0.001276545925065875,
+ "learning_rate": 3.982949361823388e-05,
+ "loss": 0.0423,
+ "step": 580
+ },
+ {
+ "epoch": 1.5526315789473686,
+ "grad_norm": 1.7920968532562256,
+ "learning_rate": 3.949110871966444e-05,
+ "loss": 0.0005,
+ "step": 590
+ },
+ {
+ "epoch": 1.5789473684210527,
+ "grad_norm": 2.7900519371032715,
+ "learning_rate": 3.914867735826489e-05,
+ "loss": 0.1829,
+ "step": 600
+ },
+ {
+ "epoch": 1.5789473684210527,
+ "eval_loss": 0.1412845253944397,
+ "eval_runtime": 1.5821,
+ "eval_samples_per_second": 50.567,
+ "eval_steps_per_second": 12.642,
+ "step": 600
+ },
+ {
+ "epoch": 1.6052631578947367,
+ "grad_norm": 4.412259578704834,
+ "learning_rate": 3.880229515375642e-05,
+ "loss": 0.0851,
+ "step": 610
+ },
+ {
+ "epoch": 1.631578947368421,
+ "grad_norm": 11.6816987991333,
+ "learning_rate": 3.845205882908432e-05,
+ "loss": 0.156,
+ "step": 620
+ },
+ {
+ "epoch": 1.6578947368421053,
+ "grad_norm": 0.01778053119778633,
+ "learning_rate": 3.809806618340922e-05,
+ "loss": 0.0431,
+ "step": 630
+ },
+ {
+ "epoch": 1.6842105263157894,
+ "grad_norm": 0.0008851743768900633,
+ "learning_rate": 3.774041606479794e-05,
+ "loss": 0.057,
+ "step": 640
+ },
+ {
+ "epoch": 1.7105263157894737,
+ "grad_norm": 0.4260232448577881,
+ "learning_rate": 3.737920834262134e-05,
+ "loss": 0.0936,
+ "step": 650
+ },
+ {
+ "epoch": 1.736842105263158,
+ "grad_norm": 0.03417075052857399,
+ "learning_rate": 3.7014543879667094e-05,
+ "loss": 0.0532,
+ "step": 660
+ },
+ {
+ "epoch": 1.763157894736842,
+ "grad_norm": 0.3643689751625061,
+ "learning_rate": 3.6646524503974955e-05,
+ "loss": 0.0547,
+ "step": 670
+ },
+ {
+ "epoch": 1.7894736842105263,
+ "grad_norm": 0.0003979972389061004,
+ "learning_rate": 3.6275252980402544e-05,
+ "loss": 0.1166,
+ "step": 680
+ },
+ {
+ "epoch": 1.8157894736842106,
+ "grad_norm": 3.4334423542022705,
+ "learning_rate": 3.590083298192957e-05,
+ "loss": 0.0327,
+ "step": 690
+ },
+ {
+ "epoch": 1.8421052631578947,
+ "grad_norm": 1.3156622648239136,
+ "learning_rate": 3.552336906070838e-05,
+ "loss": 0.0235,
+ "step": 700
+ },
+ {
+ "epoch": 1.8421052631578947,
+ "eval_loss": 0.07125674188137054,
+ "eval_runtime": 1.5769,
+ "eval_samples_per_second": 50.733,
+ "eval_steps_per_second": 12.683,
+ "step": 700
+ },
+ {
+ "epoch": 1.868421052631579,
+ "grad_norm": 0.0024085186887532473,
+ "learning_rate": 3.51429666188691e-05,
+ "loss": 0.0007,
+ "step": 710
+ },
+ {
+ "epoch": 1.8947368421052633,
+ "grad_norm": 18.228830337524414,
+ "learning_rate": 3.475973187908737e-05,
+ "loss": 0.1828,
+ "step": 720
+ },
+ {
+ "epoch": 1.9210526315789473,
+ "grad_norm": 0.008037630468606949,
+ "learning_rate": 3.437377185492303e-05,
+ "loss": 0.0011,
+ "step": 730
+ },
+ {
+ "epoch": 1.9473684210526314,
+ "grad_norm": 0.003096841275691986,
+ "learning_rate": 3.398519432093782e-05,
+ "loss": 0.1991,
+ "step": 740
+ },
+ {
+ "epoch": 1.973684210526316,
+ "grad_norm": 0.15403082966804504,
+ "learning_rate": 3.3594107782600753e-05,
+ "loss": 0.0353,
+ "step": 750
+ },
+ {
+ "epoch": 2.0,
+ "grad_norm": 7.032695770263672,
+ "learning_rate": 3.3200621445989226e-05,
+ "loss": 0.0441,
+ "step": 760
+ },
+ {
+ "epoch": 2.026315789473684,
+ "grad_norm": 0.0001874351582955569,
+ "learning_rate": 3.280484518729466e-05,
+ "loss": 0.0672,
+ "step": 770
+ },
+ {
+ "epoch": 2.0526315789473686,
+ "grad_norm": 0.000476544548291713,
+ "learning_rate": 3.2406889522140856e-05,
+ "loss": 0.0298,
+ "step": 780
+ },
+ {
+ "epoch": 2.0789473684210527,
+ "grad_norm": 0.0010313449893146753,
+ "learning_rate": 3.200686557472391e-05,
+ "loss": 0.0121,
+ "step": 790
+ },
+ {
+ "epoch": 2.1052631578947367,
+ "grad_norm": 0.0015645598759874701,
+ "learning_rate": 3.160488504678216e-05,
+ "loss": 0.1101,
+ "step": 800
+ },
+ {
+ "epoch": 2.1052631578947367,
+ "eval_loss": 0.06111542135477066,
+ "eval_runtime": 1.5827,
+ "eval_samples_per_second": 50.547,
+ "eval_steps_per_second": 12.637,
+ "step": 800
+ },
+ {
+ "epoch": 2.1052631578947367,
+ "step": 800,
+ "total_flos": 2.241573610389504e+16,
+ "train_loss": 0.32420646863698493,
+ "train_runtime": 246.1313,
+ "train_samples_per_second": 30.878,
+ "train_steps_per_second": 7.719
+ }
+ ],
+ "logging_steps": 10,
+ "max_steps": 1900,
+ "num_input_tokens_seen": 0,
+ "num_train_epochs": 5,
+ "save_steps": 100,
+ "total_flos": 2.241573610389504e+16,
+ "train_batch_size": 4,
+ "trial_name": null,
+ "trial_params": null
+}
diff --git a/llama2_13b_peft/reasoning_about_colored_objects/training_args.bin b/llama2_13b_peft/reasoning_about_colored_objects/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..277414d1964a61fad6ed583ebaa481ca797da682
--- /dev/null
+++ b/llama2_13b_peft/reasoning_about_colored_objects/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f5d412c0245eb5db7e6c9f2eb7e78eec4fdafe460bb728bd2590c7cc04591e54
+size 5240
diff --git a/llama2_13b_peft/reasoning_about_colored_objects/training_eval_loss.png b/llama2_13b_peft/reasoning_about_colored_objects/training_eval_loss.png
new file mode 100644
index 0000000000000000000000000000000000000000..39cdd7df134d366217fe008483151b17506926d3
Binary files /dev/null and b/llama2_13b_peft/reasoning_about_colored_objects/training_eval_loss.png differ
diff --git a/llama2_13b_peft/reasoning_about_colored_objects/training_loss.png b/llama2_13b_peft/reasoning_about_colored_objects/training_loss.png
new file mode 100644
index 0000000000000000000000000000000000000000..b1b4489099d97ac609f39866920d1d9404b638a5
Binary files /dev/null and b/llama2_13b_peft/reasoning_about_colored_objects/training_loss.png differ
diff --git a/llama2_13b_peft/strategyqa/README.md b/llama2_13b_peft/strategyqa/README.md
new file mode 100644
index 0000000000000000000000000000000000000000..1c7cdf93feaa413d6479ec1229832ea556b8d7f0
--- /dev/null
+++ b/llama2_13b_peft/strategyqa/README.md
@@ -0,0 +1,78 @@
+---
+license: other
+library_name: peft
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+base_model: /data1/model/llama2/meta-llama/Llama2-13b
+model-index:
+- name: strategyqa_no_sys
+ results: []
+---
+
+
+
+# strategyqa_no_sys
+
+This model is a fine-tuned version of [/data1/model/llama2/meta-llama/Llama2-13b](https://huggingface.co//data1/model/llama2/meta-llama/Llama2-13b) on the strategyqa_no_sys dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.3844
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 0.0001
+- train_batch_size: 8
+- eval_batch_size: 8
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 3
+- total_train_batch_size: 24
+- total_eval_batch_size: 24
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_steps: 20
+- num_epochs: 10.0
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 1.1108 | 0.77 | 50 | 1.1378 |
+| 1.0652 | 1.54 | 100 | 1.1236 |
+| 0.9679 | 2.31 | 150 | 1.1641 |
+| 0.7321 | 3.08 | 200 | 1.3882 |
+| 0.5562 | 3.85 | 250 | 1.3443 |
+| 0.2699 | 4.62 | 300 | 1.6109 |
+| 0.1135 | 5.38 | 350 | 1.9021 |
+| 0.0511 | 6.15 | 400 | 2.2162 |
+| 0.049 | 6.92 | 450 | 2.0983 |
+| 0.0212 | 7.69 | 500 | 2.3844 |
+| 0.0107 | 8.46 | 550 | 2.5324 |
+| 0.006 | 9.23 | 600 | 2.5657 |
+| 0.0131 | 10.0 | 650 | 2.5765 |
+
+
+### Framework versions
+
+- PEFT 0.9.0
+- Transformers 4.38.2
+- Pytorch 2.2.1
+- Datasets 2.18.0
+- Tokenizers 0.15.2
\ No newline at end of file
diff --git a/llama2_13b_peft/strategyqa/adapter_config.json b/llama2_13b_peft/strategyqa/adapter_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..2312c477123e66beb0a4a4e3ade3dc7849a6f9d8
--- /dev/null
+++ b/llama2_13b_peft/strategyqa/adapter_config.json
@@ -0,0 +1,33 @@
+{
+ "alpha_pattern": {},
+ "auto_mapping": null,
+ "base_model_name_or_path": "/data1/model/llama2/meta-llama/Llama2-13b",
+ "bias": "none",
+ "fan_in_fan_out": false,
+ "inference_mode": true,
+ "init_lora_weights": true,
+ "layers_pattern": null,
+ "layers_to_transform": null,
+ "loftq_config": {},
+ "lora_alpha": 16,
+ "lora_dropout": 0.0,
+ "megatron_config": null,
+ "megatron_core": "megatron.core",
+ "modules_to_save": null,
+ "peft_type": "LORA",
+ "r": 8,
+ "rank_pattern": {},
+ "revision": null,
+ "target_modules": [
+ "v_proj",
+ "k_proj",
+ "q_proj",
+ "gate_proj",
+ "down_proj",
+ "o_proj",
+ "up_proj"
+ ],
+ "task_type": "CAUSAL_LM",
+ "use_dora": false,
+ "use_rslora": false
+}
\ No newline at end of file
diff --git a/llama2_13b_peft/strategyqa/adapter_model.safetensors b/llama2_13b_peft/strategyqa/adapter_model.safetensors
new file mode 100644
index 0000000000000000000000000000000000000000..a09623166113b54a63753cfae50ae31d94ebf730
--- /dev/null
+++ b/llama2_13b_peft/strategyqa/adapter_model.safetensors
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc64663bb7afeb1577c54465749c48ad601484deb177e42179deade5a5fcc5f5
+size 125248064
diff --git a/llama2_13b_peft/strategyqa/all_results.json b/llama2_13b_peft/strategyqa/all_results.json
new file mode 100644
index 0000000000000000000000000000000000000000..2d9f50b0528247e1c3dead3cede5f40ee093faa0
--- /dev/null
+++ b/llama2_13b_peft/strategyqa/all_results.json
@@ -0,0 +1,11 @@
+{
+ "epoch": 10.0,
+ "eval_loss": 2.3844265937805176,
+ "eval_runtime": 1.8178,
+ "eval_samples_per_second": 151.279,
+ "eval_steps_per_second": 6.601,
+ "train_loss": 0.4327951647226627,
+ "train_runtime": 356.3752,
+ "train_samples_per_second": 43.69,
+ "train_steps_per_second": 1.824
+}
\ No newline at end of file
diff --git a/llama2_13b_peft/strategyqa/eval_results.json b/llama2_13b_peft/strategyqa/eval_results.json
new file mode 100644
index 0000000000000000000000000000000000000000..14c876bdcceb85839e9dab1b0397f5ba99e689b2
--- /dev/null
+++ b/llama2_13b_peft/strategyqa/eval_results.json
@@ -0,0 +1,7 @@
+{
+ "epoch": 10.0,
+ "eval_loss": 2.3844265937805176,
+ "eval_runtime": 1.8178,
+ "eval_samples_per_second": 151.279,
+ "eval_steps_per_second": 6.601
+}
\ No newline at end of file
diff --git a/llama2_13b_peft/strategyqa/special_tokens_map.json b/llama2_13b_peft/strategyqa/special_tokens_map.json
new file mode 100644
index 0000000000000000000000000000000000000000..72ecfeeb7e14d244c936169d2ed139eeae235ef1
--- /dev/null
+++ b/llama2_13b_peft/strategyqa/special_tokens_map.json
@@ -0,0 +1,24 @@
+{
+ "bos_token": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false
+ },
+ "eos_token": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false
+ },
+ "pad_token": "",
+ "unk_token": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false
+ }
+}
diff --git a/llama2_13b_peft/strategyqa/tokenizer.model b/llama2_13b_peft/strategyqa/tokenizer.model
new file mode 100644
index 0000000000000000000000000000000000000000..6c00c742ce03c627d6cd5b795984876fa49fa899
--- /dev/null
+++ b/llama2_13b_peft/strategyqa/tokenizer.model
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e556afd44213b6bd1be2b850ebbbd98f5481437a8021afaf58ee7fb1818d347
+size 499723
diff --git a/llama2_13b_peft/strategyqa/tokenizer_config.json b/llama2_13b_peft/strategyqa/tokenizer_config.json
new file mode 100644
index 0000000000000000000000000000000000000000..a40266f39e5b5fed14de34710d35eb9e98d6bdad
--- /dev/null
+++ b/llama2_13b_peft/strategyqa/tokenizer_config.json
@@ -0,0 +1,45 @@
+{
+ "add_bos_token": true,
+ "add_eos_token": false,
+ "add_prefix_space": true,
+ "added_tokens_decoder": {
+ "0": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false,
+ "special": true
+ },
+ "1": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false,
+ "special": true
+ },
+ "2": {
+ "content": "",
+ "lstrip": false,
+ "normalized": false,
+ "rstrip": false,
+ "single_word": false,
+ "special": true
+ }
+ },
+ "bos_token": "",
+ "chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content + '\\n' }}{% endif %}{% endfor %}",
+ "clean_up_tokenization_spaces": false,
+ "eos_token": "",
+ "legacy": true,
+ "model_max_length": 1000000000000000019884624838656,
+ "pad_token": "",
+ "padding_side": "right",
+ "sp_model_kwargs": {},
+ "spaces_between_special_tokens": false,
+ "split_special_tokens": false,
+ "tokenizer_class": "LlamaTokenizer",
+ "unk_token": "",
+ "use_default_system_prompt": false
+}
diff --git a/llama2_13b_peft/strategyqa/train_results.json b/llama2_13b_peft/strategyqa/train_results.json
new file mode 100644
index 0000000000000000000000000000000000000000..2abcbe12bda1f3a4021eaa302c5c616deb7d4190
--- /dev/null
+++ b/llama2_13b_peft/strategyqa/train_results.json
@@ -0,0 +1,7 @@
+{
+ "epoch": 10.0,
+ "train_loss": 0.4327951647226627,
+ "train_runtime": 356.3752,
+ "train_samples_per_second": 43.69,
+ "train_steps_per_second": 1.824
+}
\ No newline at end of file
diff --git a/llama2_13b_peft/strategyqa/trainer_log.jsonl b/llama2_13b_peft/strategyqa/trainer_log.jsonl
new file mode 100644
index 0000000000000000000000000000000000000000..aceb8517124217723a70d029f52fd5c42be1bc4d
--- /dev/null
+++ b/llama2_13b_peft/strategyqa/trainer_log.jsonl
@@ -0,0 +1,80 @@
+{"current_steps": 10, "total_steps": 650, "loss": 1.8151, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5e-05, "epoch": 0.15, "percentage": 1.54, "elapsed_time": "0:00:06", "remaining_time": "0:06:46"}
+{"current_steps": 20, "total_steps": 650, "loss": 1.5545, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0001, "epoch": 0.31, "percentage": 3.08, "elapsed_time": "0:00:11", "remaining_time": "0:05:51"}
+{"current_steps": 30, "total_steps": 650, "loss": 1.2226, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.993784606094612e-05, "epoch": 0.46, "percentage": 4.62, "elapsed_time": "0:00:15", "remaining_time": "0:05:29"}
+{"current_steps": 40, "total_steps": 650, "loss": 1.1648, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.975153876827008e-05, "epoch": 0.62, "percentage": 6.15, "elapsed_time": "0:00:20", "remaining_time": "0:05:16"}
+{"current_steps": 50, "total_steps": 650, "loss": 1.1108, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.944154131125642e-05, "epoch": 0.77, "percentage": 7.69, "elapsed_time": "0:00:25", "remaining_time": "0:05:07"}
+{"current_steps": 50, "total_steps": 650, "loss": null, "eval_loss": 1.1378295421600342, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 0.77, "percentage": 7.69, "elapsed_time": "0:00:25", "remaining_time": "0:05:07"}
+{"current_steps": 60, "total_steps": 650, "loss": 1.1433, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.900862439242719e-05, "epoch": 0.92, "percentage": 9.23, "elapsed_time": "0:00:32", "remaining_time": "0:05:24"}
+{"current_steps": 70, "total_steps": 650, "loss": 1.1424, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.84538643114539e-05, "epoch": 1.08, "percentage": 10.77, "elapsed_time": "0:00:38", "remaining_time": "0:05:15"}
+{"current_steps": 80, "total_steps": 650, "loss": 1.092, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.777864028930705e-05, "epoch": 1.23, "percentage": 12.31, "elapsed_time": "0:00:43", "remaining_time": "0:05:09"}
+{"current_steps": 90, "total_steps": 650, "loss": 1.0537, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.698463103929542e-05, "epoch": 1.38, "percentage": 13.85, "elapsed_time": "0:00:48", "remaining_time": "0:04:59"}
+{"current_steps": 100, "total_steps": 650, "loss": 1.0652, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.607381059352038e-05, "epoch": 1.54, "percentage": 15.38, "elapsed_time": "0:00:52", "remaining_time": "0:04:51"}
+{"current_steps": 100, "total_steps": 650, "loss": null, "eval_loss": 1.1235716342926025, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 1.54, "percentage": 15.38, "elapsed_time": "0:00:52", "remaining_time": "0:04:51"}
+{"current_steps": 110, "total_steps": 650, "loss": 1.0962, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.504844339512095e-05, "epoch": 1.69, "percentage": 16.92, "elapsed_time": "0:00:59", "remaining_time": "0:04:52"}
+{"current_steps": 120, "total_steps": 650, "loss": 1.067, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.391107866851143e-05, "epoch": 1.85, "percentage": 18.46, "elapsed_time": "0:01:04", "remaining_time": "0:04:45"}
+{"current_steps": 130, "total_steps": 650, "loss": 1.0723, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.266454408160779e-05, "epoch": 2.0, "percentage": 20.0, "elapsed_time": "0:01:09", "remaining_time": "0:04:38"}
+{"current_steps": 140, "total_steps": 650, "loss": 0.9459, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.131193871579975e-05, "epoch": 2.15, "percentage": 21.54, "elapsed_time": "0:01:14", "remaining_time": "0:04:31"}
+{"current_steps": 150, "total_steps": 650, "loss": 0.9679, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.985662536114613e-05, "epoch": 2.31, "percentage": 23.08, "elapsed_time": "0:01:19", "remaining_time": "0:04:25"}
+{"current_steps": 150, "total_steps": 650, "loss": null, "eval_loss": 1.1641343832015991, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 2.31, "percentage": 23.08, "elapsed_time": "0:01:19", "remaining_time": "0:04:25"}
+{"current_steps": 160, "total_steps": 650, "loss": 0.8452, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.83022221559489e-05, "epoch": 2.46, "percentage": 24.62, "elapsed_time": "0:01:26", "remaining_time": "0:04:24"}
+{"current_steps": 170, "total_steps": 650, "loss": 0.8922, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.665259359149132e-05, "epoch": 2.62, "percentage": 26.15, "elapsed_time": "0:01:31", "remaining_time": "0:04:17"}
+{"current_steps": 180, "total_steps": 650, "loss": 0.8441, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.491184090430364e-05, "epoch": 2.77, "percentage": 27.69, "elapsed_time": "0:01:36", "remaining_time": "0:04:11"}
+{"current_steps": 190, "total_steps": 650, "loss": 0.9151, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.308429187984297e-05, "epoch": 2.92, "percentage": 29.23, "elapsed_time": "0:01:41", "remaining_time": "0:04:05"}
+{"current_steps": 200, "total_steps": 650, "loss": 0.7321, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.117449009293668e-05, "epoch": 3.08, "percentage": 30.77, "elapsed_time": "0:01:46", "remaining_time": "0:03:59"}
+{"current_steps": 200, "total_steps": 650, "loss": null, "eval_loss": 1.3882437944412231, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 3.08, "percentage": 30.77, "elapsed_time": "0:01:46", "remaining_time": "0:03:59"}
+{"current_steps": 210, "total_steps": 650, "loss": 0.5473, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.91871836117395e-05, "epoch": 3.23, "percentage": 32.31, "elapsed_time": "0:01:53", "remaining_time": "0:03:57"}
+{"current_steps": 220, "total_steps": 650, "loss": 0.549, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.712731319328798e-05, "epoch": 3.38, "percentage": 33.85, "elapsed_time": "0:01:58", "remaining_time": "0:03:50"}
+{"current_steps": 230, "total_steps": 650, "loss": 0.5206, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.500000000000001e-05, "epoch": 3.54, "percentage": 35.38, "elapsed_time": "0:02:03", "remaining_time": "0:03:44"}
+{"current_steps": 240, "total_steps": 650, "loss": 0.5435, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.281053286765815e-05, "epoch": 3.69, "percentage": 36.92, "elapsed_time": "0:02:08", "remaining_time": "0:03:39"}
+{"current_steps": 250, "total_steps": 650, "loss": 0.5562, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.056435515653059e-05, "epoch": 3.85, "percentage": 38.46, "elapsed_time": "0:02:13", "remaining_time": "0:03:33"}
+{"current_steps": 250, "total_steps": 650, "loss": null, "eval_loss": 1.3442635536193848, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 3.85, "percentage": 38.46, "elapsed_time": "0:02:13", "remaining_time": "0:03:33"}
+{"current_steps": 260, "total_steps": 650, "loss": 0.5269, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.826705121831976e-05, "epoch": 4.0, "percentage": 40.0, "elapsed_time": "0:02:20", "remaining_time": "0:03:30"}
+{"current_steps": 270, "total_steps": 650, "loss": 0.3063, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.592433251258423e-05, "epoch": 4.15, "percentage": 41.54, "elapsed_time": "0:02:25", "remaining_time": "0:03:24"}
+{"current_steps": 280, "total_steps": 650, "loss": 0.2743, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.354202340715026e-05, "epoch": 4.31, "percentage": 43.08, "elapsed_time": "0:02:30", "remaining_time": "0:03:18"}
+{"current_steps": 290, "total_steps": 650, "loss": 0.2718, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.112604669781572e-05, "epoch": 4.46, "percentage": 44.62, "elapsed_time": "0:02:35", "remaining_time": "0:03:12"}
+{"current_steps": 300, "total_steps": 650, "loss": 0.2699, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.868240888334653e-05, "epoch": 4.62, "percentage": 46.15, "elapsed_time": "0:02:40", "remaining_time": "0:03:07"}
+{"current_steps": 300, "total_steps": 650, "loss": null, "eval_loss": 1.6108604669570923, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 4.62, "percentage": 46.15, "elapsed_time": "0:02:40", "remaining_time": "0:03:07"}
+{"current_steps": 310, "total_steps": 650, "loss": 0.2749, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.621718523237427e-05, "epoch": 4.77, "percentage": 47.69, "elapsed_time": "0:02:47", "remaining_time": "0:03:03"}
+{"current_steps": 320, "total_steps": 650, "loss": 0.2582, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.373650467932122e-05, "epoch": 4.92, "percentage": 49.23, "elapsed_time": "0:02:52", "remaining_time": "0:02:57"}
+{"current_steps": 330, "total_steps": 650, "loss": 0.2131, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.124653458690365e-05, "epoch": 5.08, "percentage": 50.77, "elapsed_time": "0:02:57", "remaining_time": "0:02:52"}
+{"current_steps": 340, "total_steps": 650, "loss": 0.1242, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.875346541309637e-05, "epoch": 5.23, "percentage": 52.31, "elapsed_time": "0:03:02", "remaining_time": "0:02:46"}
+{"current_steps": 350, "total_steps": 650, "loss": 0.1135, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.626349532067879e-05, "epoch": 5.38, "percentage": 53.85, "elapsed_time": "0:03:07", "remaining_time": "0:02:40"}
+{"current_steps": 350, "total_steps": 650, "loss": null, "eval_loss": 1.902097463607788, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 5.38, "percentage": 53.85, "elapsed_time": "0:03:07", "remaining_time": "0:02:40"}
+{"current_steps": 360, "total_steps": 650, "loss": 0.1092, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.378281476762576e-05, "epoch": 5.54, "percentage": 55.38, "elapsed_time": "0:03:13", "remaining_time": "0:02:36"}
+{"current_steps": 370, "total_steps": 650, "loss": 0.1171, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.131759111665349e-05, "epoch": 5.69, "percentage": 56.92, "elapsed_time": "0:03:19", "remaining_time": "0:02:30"}
+{"current_steps": 380, "total_steps": 650, "loss": 0.1155, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.887395330218429e-05, "epoch": 5.85, "percentage": 58.46, "elapsed_time": "0:03:23", "remaining_time": "0:02:24"}
+{"current_steps": 390, "total_steps": 650, "loss": 0.1274, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.6457976592849754e-05, "epoch": 6.0, "percentage": 60.0, "elapsed_time": "0:03:29", "remaining_time": "0:02:19"}
+{"current_steps": 400, "total_steps": 650, "loss": 0.0511, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.4075667487415785e-05, "epoch": 6.15, "percentage": 61.54, "elapsed_time": "0:03:34", "remaining_time": "0:02:13"}
+{"current_steps": 400, "total_steps": 650, "loss": null, "eval_loss": 2.216161012649536, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 6.15, "percentage": 61.54, "elapsed_time": "0:03:34", "remaining_time": "0:02:13"}
+{"current_steps": 410, "total_steps": 650, "loss": 0.0489, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.173294878168025e-05, "epoch": 6.31, "percentage": 63.08, "elapsed_time": "0:03:41", "remaining_time": "0:02:09"}
+{"current_steps": 420, "total_steps": 650, "loss": 0.0419, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.9435644843469436e-05, "epoch": 6.46, "percentage": 64.62, "elapsed_time": "0:03:45", "remaining_time": "0:02:03"}
+{"current_steps": 430, "total_steps": 650, "loss": 0.053, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.718946713234185e-05, "epoch": 6.62, "percentage": 66.15, "elapsed_time": "0:03:50", "remaining_time": "0:01:58"}
+{"current_steps": 440, "total_steps": 650, "loss": 0.0422, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.500000000000001e-05, "epoch": 6.77, "percentage": 67.69, "elapsed_time": "0:03:55", "remaining_time": "0:01:52"}
+{"current_steps": 450, "total_steps": 650, "loss": 0.049, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.2872686806712035e-05, "epoch": 6.92, "percentage": 69.23, "elapsed_time": "0:04:01", "remaining_time": "0:01:47"}
+{"current_steps": 450, "total_steps": 650, "loss": null, "eval_loss": 2.0983123779296875, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 6.92, "percentage": 69.23, "elapsed_time": "0:04:01", "remaining_time": "0:01:47"}
+{"current_steps": 460, "total_steps": 650, "loss": 0.0362, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.0812816388260518e-05, "epoch": 7.08, "percentage": 70.77, "elapsed_time": "0:04:07", "remaining_time": "0:01:42"}
+{"current_steps": 470, "total_steps": 650, "loss": 0.0207, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.8825509907063327e-05, "epoch": 7.23, "percentage": 72.31, "elapsed_time": "0:04:13", "remaining_time": "0:01:36"}
+{"current_steps": 480, "total_steps": 650, "loss": 0.0266, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.691570812015704e-05, "epoch": 7.38, "percentage": 73.85, "elapsed_time": "0:04:18", "remaining_time": "0:01:31"}
+{"current_steps": 490, "total_steps": 650, "loss": 0.0167, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.5088159095696363e-05, "epoch": 7.54, "percentage": 75.38, "elapsed_time": "0:04:23", "remaining_time": "0:01:25"}
+{"current_steps": 500, "total_steps": 650, "loss": 0.0212, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3347406408508695e-05, "epoch": 7.69, "percentage": 76.92, "elapsed_time": "0:04:28", "remaining_time": "0:01:20"}
+{"current_steps": 500, "total_steps": 650, "loss": null, "eval_loss": 2.3844265937805176, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 7.69, "percentage": 76.92, "elapsed_time": "0:04:28", "remaining_time": "0:01:20"}
+{"current_steps": 510, "total_steps": 650, "loss": 0.0186, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.1697777844051105e-05, "epoch": 7.85, "percentage": 78.46, "elapsed_time": "0:04:35", "remaining_time": "0:01:15"}
+{"current_steps": 520, "total_steps": 650, "loss": 0.0195, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0143374638853891e-05, "epoch": 8.0, "percentage": 80.0, "elapsed_time": "0:04:40", "remaining_time": "0:01:10"}
+{"current_steps": 530, "total_steps": 650, "loss": 0.0087, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.688061284200266e-06, "epoch": 8.15, "percentage": 81.54, "elapsed_time": "0:04:45", "remaining_time": "0:01:04"}
+{"current_steps": 540, "total_steps": 650, "loss": 0.0111, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.33545591839222e-06, "epoch": 8.31, "percentage": 83.08, "elapsed_time": "0:04:50", "remaining_time": "0:00:59"}
+{"current_steps": 550, "total_steps": 650, "loss": 0.0107, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.088921331488568e-06, "epoch": 8.46, "percentage": 84.62, "elapsed_time": "0:04:55", "remaining_time": "0:00:53"}
+{"current_steps": 550, "total_steps": 650, "loss": null, "eval_loss": 2.5323970317840576, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 8.46, "percentage": 84.62, "elapsed_time": "0:04:55", "remaining_time": "0:00:53"}
+{"current_steps": 560, "total_steps": 650, "loss": 0.0102, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.951556604879048e-06, "epoch": 8.62, "percentage": 86.15, "elapsed_time": "0:05:02", "remaining_time": "0:00:48"}
+{"current_steps": 570, "total_steps": 650, "loss": 0.0108, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.9261894064796135e-06, "epoch": 8.77, "percentage": 87.69, "elapsed_time": "0:05:07", "remaining_time": "0:00:43"}
+{"current_steps": 580, "total_steps": 650, "loss": 0.0105, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.0153689607045845e-06, "epoch": 8.92, "percentage": 89.23, "elapsed_time": "0:05:12", "remaining_time": "0:00:37"}
+{"current_steps": 590, "total_steps": 650, "loss": 0.0104, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.221359710692961e-06, "epoch": 9.08, "percentage": 90.77, "elapsed_time": "0:05:17", "remaining_time": "0:00:32"}
+{"current_steps": 600, "total_steps": 650, "loss": 0.006, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.5461356885461075e-06, "epoch": 9.23, "percentage": 92.31, "elapsed_time": "0:05:22", "remaining_time": "0:00:26"}
+{"current_steps": 600, "total_steps": 650, "loss": null, "eval_loss": 2.5656862258911133, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 9.23, "percentage": 92.31, "elapsed_time": "0:05:22", "remaining_time": "0:00:26"}
+{"current_steps": 610, "total_steps": 650, "loss": 0.009, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.913756075728087e-07, "epoch": 9.38, "percentage": 93.85, "elapsed_time": "0:05:29", "remaining_time": "0:00:21"}
+{"current_steps": 620, "total_steps": 650, "loss": 0.0077, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.584586887435739e-07, "epoch": 9.54, "percentage": 95.38, "elapsed_time": "0:05:34", "remaining_time": "0:00:16"}
+{"current_steps": 630, "total_steps": 650, "loss": 0.0085, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.4846123172992954e-07, "epoch": 9.69, "percentage": 96.92, "elapsed_time": "0:05:39", "remaining_time": "0:00:10"}
+{"current_steps": 640, "total_steps": 650, "loss": 0.0082, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.215393905388278e-08, "epoch": 9.85, "percentage": 98.46, "elapsed_time": "0:05:44", "remaining_time": "0:00:05"}
+{"current_steps": 650, "total_steps": 650, "loss": 0.0131, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 0.0, "epoch": 10.0, "percentage": 100.0, "elapsed_time": "0:05:49", "remaining_time": "0:00:00"}
+{"current_steps": 650, "total_steps": 650, "loss": null, "eval_loss": 2.5764715671539307, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 10.0, "percentage": 100.0, "elapsed_time": "0:05:49", "remaining_time": "0:00:00"}
+{"current_steps": 650, "total_steps": 650, "loss": null, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 10.0, "percentage": 100.0, "elapsed_time": "0:05:49", "remaining_time": "0:00:00"}
+{"current_steps": 12, "total_steps": 12, "loss": null, "eval_loss": 2.3844265937805176, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 10.0, "percentage": 100.0, "elapsed_time": "0:06:06", "remaining_time": "0:00:00"}
diff --git a/llama2_13b_peft/strategyqa/trainer_state.json b/llama2_13b_peft/strategyqa/trainer_state.json
new file mode 100644
index 0000000000000000000000000000000000000000..5f901aa00c90cf2d2a35d37fcded072159ab9bdc
--- /dev/null
+++ b/llama2_13b_peft/strategyqa/trainer_state.json
@@ -0,0 +1,589 @@
+{
+ "best_metric": 2.3844265937805176,
+ "best_model_checkpoint": "ckpt/llama2_13b_fuze15_no_sys/strategyqa_no_sys/checkpoint-500",
+ "epoch": 10.0,
+ "eval_steps": 50,
+ "global_step": 650,
+ "is_hyper_param_search": false,
+ "is_local_process_zero": true,
+ "is_world_process_zero": true,
+ "log_history": [
+ {
+ "epoch": 0.15,
+ "grad_norm": 0.40633881092071533,
+ "learning_rate": 5e-05,
+ "loss": 1.8151,
+ "step": 10
+ },
+ {
+ "epoch": 0.31,
+ "grad_norm": 0.5893375873565674,
+ "learning_rate": 0.0001,
+ "loss": 1.5545,
+ "step": 20
+ },
+ {
+ "epoch": 0.46,
+ "grad_norm": 0.4787861406803131,
+ "learning_rate": 9.993784606094612e-05,
+ "loss": 1.2226,
+ "step": 30
+ },
+ {
+ "epoch": 0.62,
+ "grad_norm": 0.5218197703361511,
+ "learning_rate": 9.975153876827008e-05,
+ "loss": 1.1648,
+ "step": 40
+ },
+ {
+ "epoch": 0.77,
+ "grad_norm": 0.4085441827774048,
+ "learning_rate": 9.944154131125642e-05,
+ "loss": 1.1108,
+ "step": 50
+ },
+ {
+ "epoch": 0.77,
+ "eval_loss": 1.1378295421600342,
+ "eval_runtime": 1.8486,
+ "eval_samples_per_second": 148.764,
+ "eval_steps_per_second": 6.492,
+ "step": 50
+ },
+ {
+ "epoch": 0.92,
+ "grad_norm": 0.38595476746559143,
+ "learning_rate": 9.900862439242719e-05,
+ "loss": 1.1433,
+ "step": 60
+ },
+ {
+ "epoch": 1.08,
+ "grad_norm": 0.38242650032043457,
+ "learning_rate": 9.84538643114539e-05,
+ "loss": 1.1424,
+ "step": 70
+ },
+ {
+ "epoch": 1.23,
+ "grad_norm": 0.36820024251937866,
+ "learning_rate": 9.777864028930705e-05,
+ "loss": 1.092,
+ "step": 80
+ },
+ {
+ "epoch": 1.38,
+ "grad_norm": 0.40036314725875854,
+ "learning_rate": 9.698463103929542e-05,
+ "loss": 1.0537,
+ "step": 90
+ },
+ {
+ "epoch": 1.54,
+ "grad_norm": 0.45688629150390625,
+ "learning_rate": 9.607381059352038e-05,
+ "loss": 1.0652,
+ "step": 100
+ },
+ {
+ "epoch": 1.54,
+ "eval_loss": 1.1235716342926025,
+ "eval_runtime": 1.8515,
+ "eval_samples_per_second": 148.526,
+ "eval_steps_per_second": 6.481,
+ "step": 100
+ },
+ {
+ "epoch": 1.69,
+ "grad_norm": 0.4719078540802002,
+ "learning_rate": 9.504844339512095e-05,
+ "loss": 1.0962,
+ "step": 110
+ },
+ {
+ "epoch": 1.85,
+ "grad_norm": 0.5598475933074951,
+ "learning_rate": 9.391107866851143e-05,
+ "loss": 1.067,
+ "step": 120
+ },
+ {
+ "epoch": 2.0,
+ "grad_norm": 0.5334481000900269,
+ "learning_rate": 9.266454408160779e-05,
+ "loss": 1.0723,
+ "step": 130
+ },
+ {
+ "epoch": 2.15,
+ "grad_norm": 0.5928361415863037,
+ "learning_rate": 9.131193871579975e-05,
+ "loss": 0.9459,
+ "step": 140
+ },
+ {
+ "epoch": 2.31,
+ "grad_norm": 0.8523222804069519,
+ "learning_rate": 8.985662536114613e-05,
+ "loss": 0.9679,
+ "step": 150
+ },
+ {
+ "epoch": 2.31,
+ "eval_loss": 1.1641343832015991,
+ "eval_runtime": 1.8561,
+ "eval_samples_per_second": 148.157,
+ "eval_steps_per_second": 6.465,
+ "step": 150
+ },
+ {
+ "epoch": 2.46,
+ "grad_norm": 0.9512267112731934,
+ "learning_rate": 8.83022221559489e-05,
+ "loss": 0.8452,
+ "step": 160
+ },
+ {
+ "epoch": 2.62,
+ "grad_norm": 0.8635448217391968,
+ "learning_rate": 8.665259359149132e-05,
+ "loss": 0.8922,
+ "step": 170
+ },
+ {
+ "epoch": 2.77,
+ "grad_norm": 1.3180458545684814,
+ "learning_rate": 8.491184090430364e-05,
+ "loss": 0.8441,
+ "step": 180
+ },
+ {
+ "epoch": 2.92,
+ "grad_norm": 1.2508196830749512,
+ "learning_rate": 8.308429187984297e-05,
+ "loss": 0.9151,
+ "step": 190
+ },
+ {
+ "epoch": 3.08,
+ "grad_norm": 1.8130565881729126,
+ "learning_rate": 8.117449009293668e-05,
+ "loss": 0.7321,
+ "step": 200
+ },
+ {
+ "epoch": 3.08,
+ "eval_loss": 1.3882437944412231,
+ "eval_runtime": 1.8575,
+ "eval_samples_per_second": 148.047,
+ "eval_steps_per_second": 6.46,
+ "step": 200
+ },
+ {
+ "epoch": 3.23,
+ "grad_norm": 1.3749483823776245,
+ "learning_rate": 7.91871836117395e-05,
+ "loss": 0.5473,
+ "step": 210
+ },
+ {
+ "epoch": 3.38,
+ "grad_norm": 2.105992078781128,
+ "learning_rate": 7.712731319328798e-05,
+ "loss": 0.549,
+ "step": 220
+ },
+ {
+ "epoch": 3.54,
+ "grad_norm": 1.8341376781463623,
+ "learning_rate": 7.500000000000001e-05,
+ "loss": 0.5206,
+ "step": 230
+ },
+ {
+ "epoch": 3.69,
+ "grad_norm": 1.7375411987304688,
+ "learning_rate": 7.281053286765815e-05,
+ "loss": 0.5435,
+ "step": 240
+ },
+ {
+ "epoch": 3.85,
+ "grad_norm": 1.8721497058868408,
+ "learning_rate": 7.056435515653059e-05,
+ "loss": 0.5562,
+ "step": 250
+ },
+ {
+ "epoch": 3.85,
+ "eval_loss": 1.3442635536193848,
+ "eval_runtime": 1.8579,
+ "eval_samples_per_second": 148.019,
+ "eval_steps_per_second": 6.459,
+ "step": 250
+ },
+ {
+ "epoch": 4.0,
+ "grad_norm": 1.7349268198013306,
+ "learning_rate": 6.826705121831976e-05,
+ "loss": 0.5269,
+ "step": 260
+ },
+ {
+ "epoch": 4.15,
+ "grad_norm": 2.820291757583618,
+ "learning_rate": 6.592433251258423e-05,
+ "loss": 0.3063,
+ "step": 270
+ },
+ {
+ "epoch": 4.31,
+ "grad_norm": 1.9900763034820557,
+ "learning_rate": 6.354202340715026e-05,
+ "loss": 0.2743,
+ "step": 280
+ },
+ {
+ "epoch": 4.46,
+ "grad_norm": 2.0194549560546875,
+ "learning_rate": 6.112604669781572e-05,
+ "loss": 0.2718,
+ "step": 290
+ },
+ {
+ "epoch": 4.62,
+ "grad_norm": 2.1579859256744385,
+ "learning_rate": 5.868240888334653e-05,
+ "loss": 0.2699,
+ "step": 300
+ },
+ {
+ "epoch": 4.62,
+ "eval_loss": 1.6108604669570923,
+ "eval_runtime": 1.858,
+ "eval_samples_per_second": 148.007,
+ "eval_steps_per_second": 6.458,
+ "step": 300
+ },
+ {
+ "epoch": 4.77,
+ "grad_norm": 1.8816920518875122,
+ "learning_rate": 5.621718523237427e-05,
+ "loss": 0.2749,
+ "step": 310
+ },
+ {
+ "epoch": 4.92,
+ "grad_norm": 1.9615857601165771,
+ "learning_rate": 5.373650467932122e-05,
+ "loss": 0.2582,
+ "step": 320
+ },
+ {
+ "epoch": 5.08,
+ "grad_norm": 1.4868065118789673,
+ "learning_rate": 5.124653458690365e-05,
+ "loss": 0.2131,
+ "step": 330
+ },
+ {
+ "epoch": 5.23,
+ "grad_norm": 2.2800121307373047,
+ "learning_rate": 4.875346541309637e-05,
+ "loss": 0.1242,
+ "step": 340
+ },
+ {
+ "epoch": 5.38,
+ "grad_norm": 1.9804664850234985,
+ "learning_rate": 4.626349532067879e-05,
+ "loss": 0.1135,
+ "step": 350
+ },
+ {
+ "epoch": 5.38,
+ "eval_loss": 1.902097463607788,
+ "eval_runtime": 1.8585,
+ "eval_samples_per_second": 147.969,
+ "eval_steps_per_second": 6.457,
+ "step": 350
+ },
+ {
+ "epoch": 5.54,
+ "grad_norm": 2.2013049125671387,
+ "learning_rate": 4.378281476762576e-05,
+ "loss": 0.1092,
+ "step": 360
+ },
+ {
+ "epoch": 5.69,
+ "grad_norm": 2.5137171745300293,
+ "learning_rate": 4.131759111665349e-05,
+ "loss": 0.1171,
+ "step": 370
+ },
+ {
+ "epoch": 5.85,
+ "grad_norm": 1.885507583618164,
+ "learning_rate": 3.887395330218429e-05,
+ "loss": 0.1155,
+ "step": 380
+ },
+ {
+ "epoch": 6.0,
+ "grad_norm": 1.9361807107925415,
+ "learning_rate": 3.6457976592849754e-05,
+ "loss": 0.1274,
+ "step": 390
+ },
+ {
+ "epoch": 6.15,
+ "grad_norm": 1.8210408687591553,
+ "learning_rate": 3.4075667487415785e-05,
+ "loss": 0.0511,
+ "step": 400
+ },
+ {
+ "epoch": 6.15,
+ "eval_loss": 2.216161012649536,
+ "eval_runtime": 1.8599,
+ "eval_samples_per_second": 147.855,
+ "eval_steps_per_second": 6.452,
+ "step": 400
+ },
+ {
+ "epoch": 6.31,
+ "grad_norm": 1.851035475730896,
+ "learning_rate": 3.173294878168025e-05,
+ "loss": 0.0489,
+ "step": 410
+ },
+ {
+ "epoch": 6.46,
+ "grad_norm": 1.3450051546096802,
+ "learning_rate": 2.9435644843469436e-05,
+ "loss": 0.0419,
+ "step": 420
+ },
+ {
+ "epoch": 6.62,
+ "grad_norm": 1.9799576997756958,
+ "learning_rate": 2.718946713234185e-05,
+ "loss": 0.053,
+ "step": 430
+ },
+ {
+ "epoch": 6.77,
+ "grad_norm": 1.1950169801712036,
+ "learning_rate": 2.500000000000001e-05,
+ "loss": 0.0422,
+ "step": 440
+ },
+ {
+ "epoch": 6.92,
+ "grad_norm": 1.1742631196975708,
+ "learning_rate": 2.2872686806712035e-05,
+ "loss": 0.049,
+ "step": 450
+ },
+ {
+ "epoch": 6.92,
+ "eval_loss": 2.0983123779296875,
+ "eval_runtime": 1.8568,
+ "eval_samples_per_second": 148.106,
+ "eval_steps_per_second": 6.463,
+ "step": 450
+ },
+ {
+ "epoch": 7.08,
+ "grad_norm": 0.7656117081642151,
+ "learning_rate": 2.0812816388260518e-05,
+ "loss": 0.0362,
+ "step": 460
+ },
+ {
+ "epoch": 7.23,
+ "grad_norm": 1.1246815919876099,
+ "learning_rate": 1.8825509907063327e-05,
+ "loss": 0.0207,
+ "step": 470
+ },
+ {
+ "epoch": 7.38,
+ "grad_norm": 0.8675357699394226,
+ "learning_rate": 1.691570812015704e-05,
+ "loss": 0.0266,
+ "step": 480
+ },
+ {
+ "epoch": 7.54,
+ "grad_norm": 0.9884489178657532,
+ "learning_rate": 1.5088159095696363e-05,
+ "loss": 0.0167,
+ "step": 490
+ },
+ {
+ "epoch": 7.69,
+ "grad_norm": 0.9568442106246948,
+ "learning_rate": 1.3347406408508695e-05,
+ "loss": 0.0212,
+ "step": 500
+ },
+ {
+ "epoch": 7.69,
+ "eval_loss": 2.3844265937805176,
+ "eval_runtime": 1.8585,
+ "eval_samples_per_second": 147.972,
+ "eval_steps_per_second": 6.457,
+ "step": 500
+ },
+ {
+ "epoch": 7.85,
+ "grad_norm": 0.8196859955787659,
+ "learning_rate": 1.1697777844051105e-05,
+ "loss": 0.0186,
+ "step": 510
+ },
+ {
+ "epoch": 8.0,
+ "grad_norm": 0.6307012438774109,
+ "learning_rate": 1.0143374638853891e-05,
+ "loss": 0.0195,
+ "step": 520
+ },
+ {
+ "epoch": 8.15,
+ "grad_norm": 0.2967863976955414,
+ "learning_rate": 8.688061284200266e-06,
+ "loss": 0.0087,
+ "step": 530
+ },
+ {
+ "epoch": 8.31,
+ "grad_norm": 0.3737642765045166,
+ "learning_rate": 7.33545591839222e-06,
+ "loss": 0.0111,
+ "step": 540
+ },
+ {
+ "epoch": 8.46,
+ "grad_norm": 0.8899824619293213,
+ "learning_rate": 6.088921331488568e-06,
+ "loss": 0.0107,
+ "step": 550
+ },
+ {
+ "epoch": 8.46,
+ "eval_loss": 2.5323970317840576,
+ "eval_runtime": 1.8601,
+ "eval_samples_per_second": 147.844,
+ "eval_steps_per_second": 6.451,
+ "step": 550
+ },
+ {
+ "epoch": 8.62,
+ "grad_norm": 0.698161244392395,
+ "learning_rate": 4.951556604879048e-06,
+ "loss": 0.0102,
+ "step": 560
+ },
+ {
+ "epoch": 8.77,
+ "grad_norm": 0.7657814025878906,
+ "learning_rate": 3.9261894064796135e-06,
+ "loss": 0.0108,
+ "step": 570
+ },
+ {
+ "epoch": 8.92,
+ "grad_norm": 0.3906441926956177,
+ "learning_rate": 3.0153689607045845e-06,
+ "loss": 0.0105,
+ "step": 580
+ },
+ {
+ "epoch": 9.08,
+ "grad_norm": 0.23798762261867523,
+ "learning_rate": 2.221359710692961e-06,
+ "loss": 0.0104,
+ "step": 590
+ },
+ {
+ "epoch": 9.23,
+ "grad_norm": 0.18085721135139465,
+ "learning_rate": 1.5461356885461075e-06,
+ "loss": 0.006,
+ "step": 600
+ },
+ {
+ "epoch": 9.23,
+ "eval_loss": 2.5656862258911133,
+ "eval_runtime": 1.8572,
+ "eval_samples_per_second": 148.074,
+ "eval_steps_per_second": 6.461,
+ "step": 600
+ },
+ {
+ "epoch": 9.38,
+ "grad_norm": 0.243088960647583,
+ "learning_rate": 9.913756075728087e-07,
+ "loss": 0.009,
+ "step": 610
+ },
+ {
+ "epoch": 9.54,
+ "grad_norm": 0.16859322786331177,
+ "learning_rate": 5.584586887435739e-07,
+ "loss": 0.0077,
+ "step": 620
+ },
+ {
+ "epoch": 9.69,
+ "grad_norm": 0.35239869356155396,
+ "learning_rate": 2.4846123172992954e-07,
+ "loss": 0.0085,
+ "step": 630
+ },
+ {
+ "epoch": 9.85,
+ "grad_norm": 0.28910598158836365,
+ "learning_rate": 6.215393905388278e-08,
+ "loss": 0.0082,
+ "step": 640
+ },
+ {
+ "epoch": 10.0,
+ "grad_norm": 0.20681537687778473,
+ "learning_rate": 0.0,
+ "loss": 0.0131,
+ "step": 650
+ },
+ {
+ "epoch": 10.0,
+ "eval_loss": 2.5764715671539307,
+ "eval_runtime": 1.8597,
+ "eval_samples_per_second": 147.872,
+ "eval_steps_per_second": 6.453,
+ "step": 650
+ },
+ {
+ "epoch": 10.0,
+ "step": 650,
+ "total_flos": 1.3867479608590336e+17,
+ "train_loss": 0.4327951647226627,
+ "train_runtime": 356.3752,
+ "train_samples_per_second": 43.69,
+ "train_steps_per_second": 1.824
+ }
+ ],
+ "logging_steps": 10,
+ "max_steps": 650,
+ "num_input_tokens_seen": 0,
+ "num_train_epochs": 10,
+ "save_steps": 500,
+ "total_flos": 1.3867479608590336e+17,
+ "train_batch_size": 8,
+ "trial_name": null,
+ "trial_params": null
+}
diff --git a/llama2_13b_peft/strategyqa/training_args.bin b/llama2_13b_peft/strategyqa/training_args.bin
new file mode 100644
index 0000000000000000000000000000000000000000..9e12846af70d08e385a22d812ce3efaf28657316
--- /dev/null
+++ b/llama2_13b_peft/strategyqa/training_args.bin
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f7d881616d890702d9bf8038c0f74d1c483ccfcc313338fc16ec81f03f4aff32
+size 5112
diff --git a/llama2_13b_peft/strategyqa/training_eval_loss.png b/llama2_13b_peft/strategyqa/training_eval_loss.png
new file mode 100644
index 0000000000000000000000000000000000000000..ffba54710c17ec7464329c4efb7cef2865fbf380
Binary files /dev/null and b/llama2_13b_peft/strategyqa/training_eval_loss.png differ
diff --git a/llama2_13b_peft/strategyqa/training_loss.png b/llama2_13b_peft/strategyqa/training_loss.png
new file mode 100644
index 0000000000000000000000000000000000000000..62edb8bbe83bdc857401d719ccd2287a7949b81c
Binary files /dev/null and b/llama2_13b_peft/strategyqa/training_loss.png differ