bogdal commited on
Commit
c23c907
1 Parent(s): 85a5914

Upload 13 files

Browse files
README.md CHANGED
@@ -1,3 +1,58 @@
1
  ---
2
- license: apache-2.0
 
 
 
 
 
 
 
 
 
3
  ---
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
+ license: other
3
+ library_name: peft
4
+ tags:
5
+ - llama-factory
6
+ - lora
7
+ - generated_from_trainer
8
+ base_model: openchat/openchat-3.5-1210
9
+ model-index:
10
+ - name: summarization_model
11
+ results: []
12
  ---
13
+
14
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
15
+ should probably proofread and complete it, then remove this comment. -->
16
+
17
+ # summarization_model
18
+
19
+ This model is a fine-tuned version of [openchat/openchat-3.5-1210](https://huggingface.co/openchat/openchat-3.5-1210) on the alpaca_gpt4_en dataset.
20
+
21
+ ## Model description
22
+
23
+ More information needed
24
+
25
+ ## Intended uses & limitations
26
+
27
+ More information needed
28
+
29
+ ## Training and evaluation data
30
+
31
+ More information needed
32
+
33
+ ## Training procedure
34
+
35
+ ### Training hyperparameters
36
+
37
+ The following hyperparameters were used during training:
38
+ - learning_rate: 5e-05
39
+ - train_batch_size: 1
40
+ - eval_batch_size: 8
41
+ - seed: 42
42
+ - gradient_accumulation_steps: 4
43
+ - total_train_batch_size: 4
44
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
45
+ - lr_scheduler_type: cosine
46
+ - num_epochs: 3.0
47
+
48
+ ### Training results
49
+
50
+
51
+
52
+ ### Framework versions
53
+
54
+ - PEFT 0.7.1
55
+ - Transformers 4.36.2
56
+ - Pytorch 2.1.2+cu121
57
+ - Datasets 2.16.1
58
+ - Tokenizers 0.15.0
adapter_config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "alpha_pattern": {},
3
+ "auto_mapping": null,
4
+ "base_model_name_or_path": "openchat/openchat-3.5-1210",
5
+ "bias": "none",
6
+ "fan_in_fan_out": false,
7
+ "inference_mode": true,
8
+ "init_lora_weights": true,
9
+ "layers_pattern": null,
10
+ "layers_to_transform": null,
11
+ "loftq_config": {},
12
+ "lora_alpha": 64,
13
+ "lora_dropout": 0.0,
14
+ "megatron_config": null,
15
+ "megatron_core": "megatron.core",
16
+ "modules_to_save": null,
17
+ "peft_type": "LORA",
18
+ "r": 32,
19
+ "rank_pattern": {},
20
+ "revision": null,
21
+ "target_modules": [
22
+ "v_proj",
23
+ "q_proj"
24
+ ],
25
+ "task_type": "CAUSAL_LM"
26
+ }
adapter_model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d857361435229532ec2ee987f31e94784c896a346d86300fd07ad424c06cde21
3
+ size 54543184
added_tokens.json ADDED
@@ -0,0 +1,4 @@
 
 
 
 
 
1
+ {
2
+ "<|end_of_turn|>": 32000,
3
+ "<|pad_0|>": 32001
4
+ }
all_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "train_loss": 0.5736952250355148,
4
+ "train_runtime": 159194.8564,
5
+ "train_samples_per_second": 0.638,
6
+ "train_steps_per_second": 0.159
7
+ }
special_tokens_map.json ADDED
@@ -0,0 +1,28 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "additional_special_tokens": [
3
+ "<|end_of_turn|>",
4
+ "<|pad_0|>"
5
+ ],
6
+ "bos_token": {
7
+ "content": "<s>",
8
+ "lstrip": false,
9
+ "normalized": false,
10
+ "rstrip": false,
11
+ "single_word": false
12
+ },
13
+ "eos_token": {
14
+ "content": "<|end_of_turn|>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false
19
+ },
20
+ "pad_token": "<|end_of_turn|>",
21
+ "unk_token": {
22
+ "content": "<unk>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false
27
+ }
28
+ }
tokenizer.model ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
3
+ size 493443
tokenizer_config.json ADDED
@@ -0,0 +1,64 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_bos_token": true,
3
+ "add_eos_token": false,
4
+ "added_tokens_decoder": {
5
+ "0": {
6
+ "content": "<unk>",
7
+ "lstrip": false,
8
+ "normalized": false,
9
+ "rstrip": false,
10
+ "single_word": false,
11
+ "special": true
12
+ },
13
+ "1": {
14
+ "content": "<s>",
15
+ "lstrip": false,
16
+ "normalized": false,
17
+ "rstrip": false,
18
+ "single_word": false,
19
+ "special": true
20
+ },
21
+ "2": {
22
+ "content": "</s>",
23
+ "lstrip": false,
24
+ "normalized": false,
25
+ "rstrip": false,
26
+ "single_word": false,
27
+ "special": true
28
+ },
29
+ "32000": {
30
+ "content": "<|end_of_turn|>",
31
+ "lstrip": false,
32
+ "normalized": false,
33
+ "rstrip": false,
34
+ "single_word": false,
35
+ "special": true
36
+ },
37
+ "32001": {
38
+ "content": "<|pad_0|>",
39
+ "lstrip": false,
40
+ "normalized": false,
41
+ "rstrip": false,
42
+ "single_word": false,
43
+ "special": true
44
+ }
45
+ },
46
+ "additional_special_tokens": [
47
+ "<|end_of_turn|>",
48
+ "<|pad_0|>"
49
+ ],
50
+ "bos_token": "<s>",
51
+ "chat_template": "{{ bos_token }}{% for message in messages %}{{ 'GPT4 Correct ' + message['role'].title() + ': ' + message['content'] + '<|end_of_turn|>'}}{% endfor %}{% if add_generation_prompt %}{{ 'GPT4 Correct Assistant:' }}{% endif %}",
52
+ "clean_up_tokenization_spaces": false,
53
+ "eos_token": "<|end_of_turn|>",
54
+ "legacy": true,
55
+ "model_max_length": 1000000000000000019884624838656,
56
+ "pad_token": "<|end_of_turn|>",
57
+ "padding_side": "right",
58
+ "sp_model_kwargs": {},
59
+ "spaces_between_special_tokens": false,
60
+ "split_special_tokens": false,
61
+ "tokenizer_class": "LlamaTokenizer",
62
+ "unk_token": "<unk>",
63
+ "use_default_system_prompt": true
64
+ }
train_results.json ADDED
@@ -0,0 +1,7 @@
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 3.0,
3
+ "train_loss": 0.5736952250355148,
4
+ "train_runtime": 159194.8564,
5
+ "train_samples_per_second": 0.638,
6
+ "train_steps_per_second": 0.159
7
+ }
trainer_log.jsonl ADDED
@@ -0,0 +1,51 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {"current_steps": 500, "total_steps": 25380, "loss": 0.7972, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9952133916247454e-05, "epoch": 0.06, "percentage": 1.97, "elapsed_time": "0:52:11", "remaining_time": "1 day, 19:16:45"}
2
+ {"current_steps": 1000, "total_steps": 25380, "loss": 0.7291, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.980871895794772e-05, "epoch": 0.12, "percentage": 3.94, "elapsed_time": "1:44:37", "remaining_time": "1 day, 18:30:44"}
3
+ {"current_steps": 1500, "total_steps": 25380, "loss": 0.7094, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.9570304302093216e-05, "epoch": 0.18, "percentage": 5.91, "elapsed_time": "2:36:46", "remaining_time": "1 day, 17:35:56"}
4
+ {"current_steps": 2000, "total_steps": 25380, "loss": 0.7107, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.923780290675475e-05, "epoch": 0.24, "percentage": 7.88, "elapsed_time": "3:29:55", "remaining_time": "1 day, 16:54:04"}
5
+ {"current_steps": 2500, "total_steps": 25380, "loss": 0.6958, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.881248801510328e-05, "epoch": 0.3, "percentage": 9.85, "elapsed_time": "4:22:11", "remaining_time": "1 day, 15:59:38"}
6
+ {"current_steps": 3000, "total_steps": 25380, "loss": 0.6906, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.829598827979682e-05, "epoch": 0.35, "percentage": 11.82, "elapsed_time": "5:13:20", "remaining_time": "1 day, 14:57:30"}
7
+ {"current_steps": 3500, "total_steps": 25380, "loss": 0.6803, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.7690281526402436e-05, "epoch": 0.41, "percentage": 13.79, "elapsed_time": "6:06:05", "remaining_time": "1 day, 14:08:36"}
8
+ {"current_steps": 4000, "total_steps": 25380, "loss": 0.6902, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.699768717973511e-05, "epoch": 0.47, "percentage": 15.76, "elapsed_time": "6:58:41", "remaining_time": "1 day, 13:17:52"}
9
+ {"current_steps": 4500, "total_steps": 25380, "loss": 0.671, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.622085738211518e-05, "epoch": 0.53, "percentage": 17.73, "elapsed_time": "7:50:06", "remaining_time": "1 day, 12:21:17"}
10
+ {"current_steps": 5000, "total_steps": 25380, "loss": 0.6747, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.536276683755501e-05, "epoch": 0.59, "percentage": 19.7, "elapsed_time": "8:41:15", "remaining_time": "1 day, 11:24:39"}
11
+ {"current_steps": 5500, "total_steps": 25380, "loss": 0.6682, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.442670142076442e-05, "epoch": 0.65, "percentage": 21.67, "elapsed_time": "9:33:38", "remaining_time": "1 day, 10:33:26"}
12
+ {"current_steps": 6000, "total_steps": 25380, "loss": 0.6545, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.341624559459447e-05, "epoch": 0.71, "percentage": 23.64, "elapsed_time": "10:25:37", "remaining_time": "1 day, 9:40:46"}
13
+ {"current_steps": 6500, "total_steps": 25380, "loss": 0.6574, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.233526868410146e-05, "epoch": 0.77, "percentage": 25.61, "elapsed_time": "11:18:28", "remaining_time": "1 day, 8:50:41"}
14
+ {"current_steps": 7000, "total_steps": 25380, "loss": 0.6623, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.1187910059791954e-05, "epoch": 0.83, "percentage": 27.58, "elapsed_time": "12:10:48", "remaining_time": "1 day, 7:58:53"}
15
+ {"current_steps": 7500, "total_steps": 25380, "loss": 0.6684, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.99785632867864e-05, "epoch": 0.89, "percentage": 29.55, "elapsed_time": "13:03:48", "remaining_time": "1 day, 7:08:35"}
16
+ {"current_steps": 8000, "total_steps": 25380, "loss": 0.6489, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.871185930059859e-05, "epoch": 0.95, "percentage": 31.52, "elapsed_time": "13:55:37", "remaining_time": "1 day, 6:15:23"}
17
+ {"current_steps": 8500, "total_steps": 25380, "loss": 0.6594, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.739264867395593e-05, "epoch": 1.0, "percentage": 33.49, "elapsed_time": "14:47:45", "remaining_time": "1 day, 5:22:58"}
18
+ {"current_steps": 9000, "total_steps": 25380, "loss": 0.564, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.6025983042565795e-05, "epoch": 1.06, "percentage": 35.46, "elapsed_time": "15:39:23", "remaining_time": "1 day, 4:29:41"}
19
+ {"current_steps": 9500, "total_steps": 25380, "loss": 0.577, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.461709576095409e-05, "epoch": 1.12, "percentage": 37.43, "elapsed_time": "16:32:26", "remaining_time": "1 day, 3:38:56"}
20
+ {"current_steps": 10000, "total_steps": 25380, "loss": 0.5692, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.317138186245037e-05, "epoch": 1.18, "percentage": 39.4, "elapsed_time": "17:24:44", "remaining_time": "1 day, 2:46:49"}
21
+ {"current_steps": 10500, "total_steps": 25380, "loss": 0.5748, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.169437740005849e-05, "epoch": 1.24, "percentage": 41.37, "elapsed_time": "18:17:25", "remaining_time": "1 day, 1:55:11"}
22
+ {"current_steps": 11000, "total_steps": 25380, "loss": 0.5665, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.0191738247322414e-05, "epoch": 1.3, "percentage": 43.34, "elapsed_time": "19:10:03", "remaining_time": "1 day, 1:03:26"}
23
+ {"current_steps": 11500, "total_steps": 25380, "loss": 0.5752, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.8669218440364937e-05, "epoch": 1.36, "percentage": 45.31, "elapsed_time": "20:02:59", "remaining_time": "1 day, 0:11:57"}
24
+ {"current_steps": 12000, "total_steps": 25380, "loss": 0.5591, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.713264814403362e-05, "epoch": 1.42, "percentage": 47.28, "elapsed_time": "20:55:23", "remaining_time": "23:19:45"}
25
+ {"current_steps": 12500, "total_steps": 25380, "loss": 0.5618, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.5587911326528148e-05, "epoch": 1.48, "percentage": 49.25, "elapsed_time": "21:47:47", "remaining_time": "22:27:32"}
26
+ {"current_steps": 13000, "total_steps": 25380, "loss": 0.5586, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.4040923227999103e-05, "epoch": 1.54, "percentage": 51.22, "elapsed_time": "22:39:30", "remaining_time": "21:34:40"}
27
+ {"current_steps": 13500, "total_steps": 25380, "loss": 0.5674, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.2497607709397543e-05, "epoch": 1.6, "percentage": 53.19, "elapsed_time": "23:32:30", "remaining_time": "20:43:00"}
28
+ {"current_steps": 14000, "total_steps": 25380, "loss": 0.5704, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.096387456831309e-05, "epoch": 1.65, "percentage": 55.16, "elapsed_time": "1 day, 0:24:15", "remaining_time": "19:50:14"}
29
+ {"current_steps": 14500, "total_steps": 25380, "loss": 0.5632, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.944559690866454e-05, "epoch": 1.71, "percentage": 57.13, "elapsed_time": "1 day, 1:16:26", "remaining_time": "18:57:51"}
30
+ {"current_steps": 15000, "total_steps": 25380, "loss": 0.5693, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.794858865090123e-05, "epoch": 1.77, "percentage": 59.1, "elapsed_time": "1 day, 2:08:26", "remaining_time": "18:05:22"}
31
+ {"current_steps": 15500, "total_steps": 25380, "loss": 0.5668, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.6478582268834674e-05, "epoch": 1.83, "percentage": 61.07, "elapsed_time": "1 day, 3:00:14", "remaining_time": "17:12:46"}
32
+ {"current_steps": 16000, "total_steps": 25380, "loss": 0.5657, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.5041206838352956e-05, "epoch": 1.89, "percentage": 63.04, "elapsed_time": "1 day, 3:52:30", "remaining_time": "16:20:30"}
33
+ {"current_steps": 16500, "total_steps": 25380, "loss": 0.5499, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.3641966482075208e-05, "epoch": 1.95, "percentage": 65.01, "elapsed_time": "1 day, 4:44:37", "remaining_time": "15:28:09"}
34
+ {"current_steps": 17000, "total_steps": 25380, "loss": 0.5471, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.228621929248813e-05, "epoch": 2.01, "percentage": 66.98, "elapsed_time": "1 day, 5:37:26", "remaining_time": "14:36:10"}
35
+ {"current_steps": 17500, "total_steps": 25380, "loss": 0.4683, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0979156814273622e-05, "epoch": 2.07, "percentage": 68.95, "elapsed_time": "1 day, 6:29:38", "remaining_time": "13:43:51"}
36
+ {"current_steps": 18000, "total_steps": 25380, "loss": 0.468, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 9.72578416439587e-06, "epoch": 2.13, "percentage": 70.92, "elapsed_time": "1 day, 7:21:55", "remaining_time": "12:51:35"}
37
+ {"current_steps": 18500, "total_steps": 25380, "loss": 0.4757, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 8.530900866073433e-06, "epoch": 2.19, "percentage": 72.89, "elapsed_time": "1 day, 8:14:38", "remaining_time": "11:59:28"}
38
+ {"current_steps": 19000, "total_steps": 25380, "loss": 0.472, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 7.399082470028884e-06, "epoch": 2.25, "percentage": 74.86, "elapsed_time": "1 day, 9:07:22", "remaining_time": "11:07:20"}
39
+ {"current_steps": 19500, "total_steps": 25380, "loss": 0.4652, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.334663033393229e-06, "epoch": 2.3, "percentage": 76.83, "elapsed_time": "1 day, 9:58:58", "remaining_time": "10:14:49"}
40
+ {"current_steps": 20000, "total_steps": 25380, "loss": 0.4665, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 5.341718523358624e-06, "epoch": 2.36, "percentage": 78.8, "elapsed_time": "1 day, 10:51:07", "remaining_time": "9:22:30"}
41
+ {"current_steps": 20500, "total_steps": 25380, "loss": 0.4719, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 4.4240512091313994e-06, "epoch": 2.42, "percentage": 80.77, "elapsed_time": "1 day, 11:43:22", "remaining_time": "8:30:13"}
42
+ {"current_steps": 21000, "total_steps": 25380, "loss": 0.4658, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.5851751019531088e-06, "epoch": 2.48, "percentage": 82.74, "elapsed_time": "1 day, 12:35:10", "remaining_time": "7:37:51"}
43
+ {"current_steps": 21500, "total_steps": 25380, "loss": 0.471, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.828302498944102e-06, "epoch": 2.54, "percentage": 84.71, "elapsed_time": "1 day, 13:27:35", "remaining_time": "6:45:36"}
44
+ {"current_steps": 22000, "total_steps": 25380, "loss": 0.4701, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.1563316822968333e-06, "epoch": 2.6, "percentage": 86.68, "elapsed_time": "1 day, 14:19:23", "remaining_time": "5:53:16"}
45
+ {"current_steps": 22500, "total_steps": 25380, "loss": 0.4681, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.5718358209224153e-06, "epoch": 2.66, "percentage": 88.65, "elapsed_time": "1 day, 15:12:39", "remaining_time": "5:01:08"}
46
+ {"current_steps": 23000, "total_steps": 25380, "loss": 0.4615, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.0770531170491289e-06, "epoch": 2.72, "percentage": 90.62, "elapsed_time": "1 day, 16:04:47", "remaining_time": "4:08:50"}
47
+ {"current_steps": 23500, "total_steps": 25380, "loss": 0.4687, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 6.738782355044049e-07, "epoch": 2.78, "percentage": 92.59, "elapsed_time": "1 day, 16:56:44", "remaining_time": "3:16:32"}
48
+ {"current_steps": 24000, "total_steps": 25380, "loss": 0.4629, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 3.638550485000031e-07, "epoch": 2.84, "percentage": 94.56, "elapsed_time": "1 day, 17:48:43", "remaining_time": "2:24:15"}
49
+ {"current_steps": 24500, "total_steps": 25380, "loss": 0.463, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 1.481707237026758e-07, "epoch": 2.9, "percentage": 96.53, "elapsed_time": "1 day, 18:41:18", "remaining_time": "1:31:59"}
50
+ {"current_steps": 25000, "total_steps": 25380, "loss": 0.4716, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": 2.7651178228807473e-08, "epoch": 2.95, "percentage": 98.5, "elapsed_time": "1 day, 19:33:52", "remaining_time": "0:39:43"}
51
+ {"current_steps": 25380, "total_steps": 25380, "loss": null, "eval_loss": null, "predict_loss": null, "reward": null, "learning_rate": null, "epoch": 3.0, "percentage": 100.0, "elapsed_time": "1 day, 20:13:14", "remaining_time": "0:00:00"}
trainer_state.json ADDED
@@ -0,0 +1,330 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 2.999734066128889,
5
+ "eval_steps": 500,
6
+ "global_step": 25380,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 0.06,
13
+ "learning_rate": 4.9952133916247454e-05,
14
+ "loss": 0.7972,
15
+ "step": 500
16
+ },
17
+ {
18
+ "epoch": 0.12,
19
+ "learning_rate": 4.980871895794772e-05,
20
+ "loss": 0.7291,
21
+ "step": 1000
22
+ },
23
+ {
24
+ "epoch": 0.18,
25
+ "learning_rate": 4.9570304302093216e-05,
26
+ "loss": 0.7094,
27
+ "step": 1500
28
+ },
29
+ {
30
+ "epoch": 0.24,
31
+ "learning_rate": 4.923780290675475e-05,
32
+ "loss": 0.7107,
33
+ "step": 2000
34
+ },
35
+ {
36
+ "epoch": 0.3,
37
+ "learning_rate": 4.881248801510328e-05,
38
+ "loss": 0.6958,
39
+ "step": 2500
40
+ },
41
+ {
42
+ "epoch": 0.35,
43
+ "learning_rate": 4.829598827979682e-05,
44
+ "loss": 0.6906,
45
+ "step": 3000
46
+ },
47
+ {
48
+ "epoch": 0.41,
49
+ "learning_rate": 4.7690281526402436e-05,
50
+ "loss": 0.6803,
51
+ "step": 3500
52
+ },
53
+ {
54
+ "epoch": 0.47,
55
+ "learning_rate": 4.699768717973511e-05,
56
+ "loss": 0.6902,
57
+ "step": 4000
58
+ },
59
+ {
60
+ "epoch": 0.53,
61
+ "learning_rate": 4.622085738211518e-05,
62
+ "loss": 0.671,
63
+ "step": 4500
64
+ },
65
+ {
66
+ "epoch": 0.59,
67
+ "learning_rate": 4.536276683755501e-05,
68
+ "loss": 0.6747,
69
+ "step": 5000
70
+ },
71
+ {
72
+ "epoch": 0.65,
73
+ "learning_rate": 4.442670142076442e-05,
74
+ "loss": 0.6682,
75
+ "step": 5500
76
+ },
77
+ {
78
+ "epoch": 0.71,
79
+ "learning_rate": 4.341624559459447e-05,
80
+ "loss": 0.6545,
81
+ "step": 6000
82
+ },
83
+ {
84
+ "epoch": 0.77,
85
+ "learning_rate": 4.233526868410146e-05,
86
+ "loss": 0.6574,
87
+ "step": 6500
88
+ },
89
+ {
90
+ "epoch": 0.83,
91
+ "learning_rate": 4.1187910059791954e-05,
92
+ "loss": 0.6623,
93
+ "step": 7000
94
+ },
95
+ {
96
+ "epoch": 0.89,
97
+ "learning_rate": 3.99785632867864e-05,
98
+ "loss": 0.6684,
99
+ "step": 7500
100
+ },
101
+ {
102
+ "epoch": 0.95,
103
+ "learning_rate": 3.871185930059859e-05,
104
+ "loss": 0.6489,
105
+ "step": 8000
106
+ },
107
+ {
108
+ "epoch": 1.0,
109
+ "learning_rate": 3.739264867395593e-05,
110
+ "loss": 0.6594,
111
+ "step": 8500
112
+ },
113
+ {
114
+ "epoch": 1.06,
115
+ "learning_rate": 3.6025983042565795e-05,
116
+ "loss": 0.564,
117
+ "step": 9000
118
+ },
119
+ {
120
+ "epoch": 1.12,
121
+ "learning_rate": 3.461709576095409e-05,
122
+ "loss": 0.577,
123
+ "step": 9500
124
+ },
125
+ {
126
+ "epoch": 1.18,
127
+ "learning_rate": 3.317138186245037e-05,
128
+ "loss": 0.5692,
129
+ "step": 10000
130
+ },
131
+ {
132
+ "epoch": 1.24,
133
+ "learning_rate": 3.169437740005849e-05,
134
+ "loss": 0.5748,
135
+ "step": 10500
136
+ },
137
+ {
138
+ "epoch": 1.3,
139
+ "learning_rate": 3.0191738247322414e-05,
140
+ "loss": 0.5665,
141
+ "step": 11000
142
+ },
143
+ {
144
+ "epoch": 1.36,
145
+ "learning_rate": 2.8669218440364937e-05,
146
+ "loss": 0.5752,
147
+ "step": 11500
148
+ },
149
+ {
150
+ "epoch": 1.42,
151
+ "learning_rate": 2.713264814403362e-05,
152
+ "loss": 0.5591,
153
+ "step": 12000
154
+ },
155
+ {
156
+ "epoch": 1.48,
157
+ "learning_rate": 2.5587911326528148e-05,
158
+ "loss": 0.5618,
159
+ "step": 12500
160
+ },
161
+ {
162
+ "epoch": 1.54,
163
+ "learning_rate": 2.4040923227999103e-05,
164
+ "loss": 0.5586,
165
+ "step": 13000
166
+ },
167
+ {
168
+ "epoch": 1.6,
169
+ "learning_rate": 2.2497607709397543e-05,
170
+ "loss": 0.5674,
171
+ "step": 13500
172
+ },
173
+ {
174
+ "epoch": 1.65,
175
+ "learning_rate": 2.096387456831309e-05,
176
+ "loss": 0.5704,
177
+ "step": 14000
178
+ },
179
+ {
180
+ "epoch": 1.71,
181
+ "learning_rate": 1.944559690866454e-05,
182
+ "loss": 0.5632,
183
+ "step": 14500
184
+ },
185
+ {
186
+ "epoch": 1.77,
187
+ "learning_rate": 1.794858865090123e-05,
188
+ "loss": 0.5693,
189
+ "step": 15000
190
+ },
191
+ {
192
+ "epoch": 1.83,
193
+ "learning_rate": 1.6478582268834674e-05,
194
+ "loss": 0.5668,
195
+ "step": 15500
196
+ },
197
+ {
198
+ "epoch": 1.89,
199
+ "learning_rate": 1.5041206838352956e-05,
200
+ "loss": 0.5657,
201
+ "step": 16000
202
+ },
203
+ {
204
+ "epoch": 1.95,
205
+ "learning_rate": 1.3641966482075208e-05,
206
+ "loss": 0.5499,
207
+ "step": 16500
208
+ },
209
+ {
210
+ "epoch": 2.01,
211
+ "learning_rate": 1.228621929248813e-05,
212
+ "loss": 0.5471,
213
+ "step": 17000
214
+ },
215
+ {
216
+ "epoch": 2.07,
217
+ "learning_rate": 1.0979156814273622e-05,
218
+ "loss": 0.4683,
219
+ "step": 17500
220
+ },
221
+ {
222
+ "epoch": 2.13,
223
+ "learning_rate": 9.72578416439587e-06,
224
+ "loss": 0.468,
225
+ "step": 18000
226
+ },
227
+ {
228
+ "epoch": 2.19,
229
+ "learning_rate": 8.530900866073433e-06,
230
+ "loss": 0.4757,
231
+ "step": 18500
232
+ },
233
+ {
234
+ "epoch": 2.25,
235
+ "learning_rate": 7.399082470028884e-06,
236
+ "loss": 0.472,
237
+ "step": 19000
238
+ },
239
+ {
240
+ "epoch": 2.3,
241
+ "learning_rate": 6.334663033393229e-06,
242
+ "loss": 0.4652,
243
+ "step": 19500
244
+ },
245
+ {
246
+ "epoch": 2.36,
247
+ "learning_rate": 5.341718523358624e-06,
248
+ "loss": 0.4665,
249
+ "step": 20000
250
+ },
251
+ {
252
+ "epoch": 2.42,
253
+ "learning_rate": 4.4240512091313994e-06,
254
+ "loss": 0.4719,
255
+ "step": 20500
256
+ },
257
+ {
258
+ "epoch": 2.48,
259
+ "learning_rate": 3.5851751019531088e-06,
260
+ "loss": 0.4658,
261
+ "step": 21000
262
+ },
263
+ {
264
+ "epoch": 2.54,
265
+ "learning_rate": 2.828302498944102e-06,
266
+ "loss": 0.471,
267
+ "step": 21500
268
+ },
269
+ {
270
+ "epoch": 2.6,
271
+ "learning_rate": 2.1563316822968333e-06,
272
+ "loss": 0.4701,
273
+ "step": 22000
274
+ },
275
+ {
276
+ "epoch": 2.66,
277
+ "learning_rate": 1.5718358209224153e-06,
278
+ "loss": 0.4681,
279
+ "step": 22500
280
+ },
281
+ {
282
+ "epoch": 2.72,
283
+ "learning_rate": 1.0770531170491289e-06,
284
+ "loss": 0.4615,
285
+ "step": 23000
286
+ },
287
+ {
288
+ "epoch": 2.78,
289
+ "learning_rate": 6.738782355044049e-07,
290
+ "loss": 0.4687,
291
+ "step": 23500
292
+ },
293
+ {
294
+ "epoch": 2.84,
295
+ "learning_rate": 3.638550485000031e-07,
296
+ "loss": 0.4629,
297
+ "step": 24000
298
+ },
299
+ {
300
+ "epoch": 2.9,
301
+ "learning_rate": 1.481707237026758e-07,
302
+ "loss": 0.463,
303
+ "step": 24500
304
+ },
305
+ {
306
+ "epoch": 2.95,
307
+ "learning_rate": 2.7651178228807473e-08,
308
+ "loss": 0.4716,
309
+ "step": 25000
310
+ },
311
+ {
312
+ "epoch": 3.0,
313
+ "step": 25380,
314
+ "total_flos": 1.6998673876237025e+18,
315
+ "train_loss": 0.5736952250355148,
316
+ "train_runtime": 159194.8564,
317
+ "train_samples_per_second": 0.638,
318
+ "train_steps_per_second": 0.159
319
+ }
320
+ ],
321
+ "logging_steps": 500,
322
+ "max_steps": 25380,
323
+ "num_input_tokens_seen": 0,
324
+ "num_train_epochs": 3,
325
+ "save_steps": 500,
326
+ "total_flos": 1.6998673876237025e+18,
327
+ "train_batch_size": 1,
328
+ "trial_name": null,
329
+ "trial_params": null
330
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0888a1d09a7d2129b9246a5ad9fbf5eed36584ffcd0ce38c5ad5f815d4e65ed
3
+ size 4856
training_loss.png ADDED