End of training
Browse files- README.md +70 -0
- adapter_1/adapter_config.json +32 -0
- adapter_1/adapter_model.safetensors +3 -0
- adapter_config.json +32 -0
- adapter_model.safetensors +3 -0
- preprocessor_config.json +14 -0
- trainer_state.json +1636 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,70 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
library_name: peft
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
base_model: openai/whisper-small
|
| 7 |
+
model-index:
|
| 8 |
+
- name: Whisper-small-Ar-MDD
|
| 9 |
+
results: []
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 13 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 14 |
+
|
| 15 |
+
# Whisper-small-Ar-MDD
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [openai/whisper-small](https://huggingface.co/openai/whisper-small) on an unknown dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.2137
|
| 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: 0.001
|
| 39 |
+
- train_batch_size: 6
|
| 40 |
+
- eval_batch_size: 6
|
| 41 |
+
- seed: 42
|
| 42 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 43 |
+
- lr_scheduler_type: linear
|
| 44 |
+
- lr_scheduler_warmup_steps: 50
|
| 45 |
+
- num_epochs: 10
|
| 46 |
+
- mixed_precision_training: Native AMP
|
| 47 |
+
|
| 48 |
+
### Training results
|
| 49 |
+
|
| 50 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 51 |
+
|:-------------:|:-----:|:----:|:---------------:|
|
| 52 |
+
| 0.0681 | 1.0 | 546 | 0.1955 |
|
| 53 |
+
| 0.0367 | 2.0 | 1092 | 0.1992 |
|
| 54 |
+
| 0.0382 | 3.0 | 1638 | 0.1857 |
|
| 55 |
+
| 0.0189 | 4.0 | 2184 | 0.1970 |
|
| 56 |
+
| 0.0274 | 5.0 | 2730 | 0.1894 |
|
| 57 |
+
| 0.02 | 6.0 | 3276 | 0.1877 |
|
| 58 |
+
| 0.0087 | 7.0 | 3822 | 0.1908 |
|
| 59 |
+
| 0.0066 | 8.0 | 4368 | 0.2085 |
|
| 60 |
+
| 0.0055 | 9.0 | 4914 | 0.2100 |
|
| 61 |
+
| 0.0013 | 10.0 | 5460 | 0.2137 |
|
| 62 |
+
|
| 63 |
+
|
| 64 |
+
### Framework versions
|
| 65 |
+
|
| 66 |
+
- PEFT 0.10.0
|
| 67 |
+
- Transformers 4.38.2
|
| 68 |
+
- Pytorch 2.2.1+cu121
|
| 69 |
+
- Datasets 2.18.0
|
| 70 |
+
- Tokenizers 0.15.2
|
adapter_1/adapter_config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": {
|
| 4 |
+
"base_model_class": "WhisperForConditionalGeneration",
|
| 5 |
+
"parent_library": "transformers.models.whisper.modeling_whisper"
|
| 6 |
+
},
|
| 7 |
+
"base_model_name_or_path": "openai/whisper-small",
|
| 8 |
+
"bias": "none",
|
| 9 |
+
"fan_in_fan_out": false,
|
| 10 |
+
"inference_mode": true,
|
| 11 |
+
"init_lora_weights": true,
|
| 12 |
+
"layer_replication": null,
|
| 13 |
+
"layers_pattern": null,
|
| 14 |
+
"layers_to_transform": null,
|
| 15 |
+
"loftq_config": {},
|
| 16 |
+
"lora_alpha": 64,
|
| 17 |
+
"lora_dropout": 0.05,
|
| 18 |
+
"megatron_config": null,
|
| 19 |
+
"megatron_core": "megatron.core",
|
| 20 |
+
"modules_to_save": null,
|
| 21 |
+
"peft_type": "LORA",
|
| 22 |
+
"r": 32,
|
| 23 |
+
"rank_pattern": {},
|
| 24 |
+
"revision": null,
|
| 25 |
+
"target_modules": [
|
| 26 |
+
"q_proj",
|
| 27 |
+
"v_proj"
|
| 28 |
+
],
|
| 29 |
+
"task_type": null,
|
| 30 |
+
"use_dora": false,
|
| 31 |
+
"use_rslora": false
|
| 32 |
+
}
|
adapter_1/adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:abb0ebdc47891ae516deaeab040b653abd88b2dceb9990155159ce05013d93b9
|
| 3 |
+
size 14176064
|
adapter_config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": {
|
| 4 |
+
"base_model_class": "WhisperForConditionalGeneration",
|
| 5 |
+
"parent_library": "transformers.models.whisper.modeling_whisper"
|
| 6 |
+
},
|
| 7 |
+
"base_model_name_or_path": "openai/whisper-small",
|
| 8 |
+
"bias": "none",
|
| 9 |
+
"fan_in_fan_out": false,
|
| 10 |
+
"inference_mode": true,
|
| 11 |
+
"init_lora_weights": true,
|
| 12 |
+
"layer_replication": null,
|
| 13 |
+
"layers_pattern": null,
|
| 14 |
+
"layers_to_transform": null,
|
| 15 |
+
"loftq_config": {},
|
| 16 |
+
"lora_alpha": 64,
|
| 17 |
+
"lora_dropout": 0.05,
|
| 18 |
+
"megatron_config": null,
|
| 19 |
+
"megatron_core": "megatron.core",
|
| 20 |
+
"modules_to_save": null,
|
| 21 |
+
"peft_type": "LORA",
|
| 22 |
+
"r": 32,
|
| 23 |
+
"rank_pattern": {},
|
| 24 |
+
"revision": null,
|
| 25 |
+
"target_modules": [
|
| 26 |
+
"q_proj",
|
| 27 |
+
"v_proj"
|
| 28 |
+
],
|
| 29 |
+
"task_type": null,
|
| 30 |
+
"use_dora": false,
|
| 31 |
+
"use_rslora": false
|
| 32 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:de7195721032862164c68a850217a7bbc5a0df6dd26266eb5ee8c195bfb57721
|
| 3 |
+
size 14176064
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"chunk_length": 30,
|
| 3 |
+
"feature_extractor_type": "WhisperFeatureExtractor",
|
| 4 |
+
"feature_size": 80,
|
| 5 |
+
"hop_length": 160,
|
| 6 |
+
"n_fft": 400,
|
| 7 |
+
"n_samples": 480000,
|
| 8 |
+
"nb_max_frames": 3000,
|
| 9 |
+
"padding_side": "right",
|
| 10 |
+
"padding_value": 0.0,
|
| 11 |
+
"processor_class": "WhisperProcessor",
|
| 12 |
+
"return_attention_mask": false,
|
| 13 |
+
"sampling_rate": 16000
|
| 14 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1636 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 5460,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.05,
|
| 13 |
+
"grad_norm": Infinity,
|
| 14 |
+
"learning_rate": 0.00044,
|
| 15 |
+
"loss": 4.1397,
|
| 16 |
+
"step": 25
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.09,
|
| 20 |
+
"grad_norm": 0.8320172429084778,
|
| 21 |
+
"learning_rate": 0.00094,
|
| 22 |
+
"loss": 1.0739,
|
| 23 |
+
"step": 50
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.14,
|
| 27 |
+
"grad_norm": 1.5708420276641846,
|
| 28 |
+
"learning_rate": 0.0009959334565619224,
|
| 29 |
+
"loss": 0.4997,
|
| 30 |
+
"step": 75
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.18,
|
| 34 |
+
"grad_norm": 0.5893439650535583,
|
| 35 |
+
"learning_rate": 0.000991312384473198,
|
| 36 |
+
"loss": 0.1246,
|
| 37 |
+
"step": 100
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.23,
|
| 41 |
+
"grad_norm": 2.0901906490325928,
|
| 42 |
+
"learning_rate": 0.0009866913123844732,
|
| 43 |
+
"loss": 0.1273,
|
| 44 |
+
"step": 125
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.27,
|
| 48 |
+
"grad_norm": 0.9590554237365723,
|
| 49 |
+
"learning_rate": 0.0009820702402957486,
|
| 50 |
+
"loss": 0.1257,
|
| 51 |
+
"step": 150
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.32,
|
| 55 |
+
"grad_norm": 1.531774878501892,
|
| 56 |
+
"learning_rate": 0.000977449168207024,
|
| 57 |
+
"loss": 0.1024,
|
| 58 |
+
"step": 175
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.37,
|
| 62 |
+
"grad_norm": 1.2284561395645142,
|
| 63 |
+
"learning_rate": 0.0009728280961182994,
|
| 64 |
+
"loss": 0.1041,
|
| 65 |
+
"step": 200
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.41,
|
| 69 |
+
"grad_norm": 1.0752886533737183,
|
| 70 |
+
"learning_rate": 0.0009682070240295749,
|
| 71 |
+
"loss": 0.1225,
|
| 72 |
+
"step": 225
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.46,
|
| 76 |
+
"grad_norm": 0.7119214534759521,
|
| 77 |
+
"learning_rate": 0.0009635859519408503,
|
| 78 |
+
"loss": 0.0904,
|
| 79 |
+
"step": 250
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.5,
|
| 83 |
+
"grad_norm": 1.0164552927017212,
|
| 84 |
+
"learning_rate": 0.0009589648798521257,
|
| 85 |
+
"loss": 0.0773,
|
| 86 |
+
"step": 275
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.55,
|
| 90 |
+
"grad_norm": 0.4222384989261627,
|
| 91 |
+
"learning_rate": 0.0009543438077634012,
|
| 92 |
+
"loss": 0.2081,
|
| 93 |
+
"step": 300
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.6,
|
| 97 |
+
"grad_norm": 0.17551083862781525,
|
| 98 |
+
"learning_rate": 0.0009497227356746766,
|
| 99 |
+
"loss": 0.0863,
|
| 100 |
+
"step": 325
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.64,
|
| 104 |
+
"grad_norm": 0.41848981380462646,
|
| 105 |
+
"learning_rate": 0.000945101663585952,
|
| 106 |
+
"loss": 0.0632,
|
| 107 |
+
"step": 350
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.69,
|
| 111 |
+
"grad_norm": 0.7539293766021729,
|
| 112 |
+
"learning_rate": 0.0009404805914972274,
|
| 113 |
+
"loss": 0.077,
|
| 114 |
+
"step": 375
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.73,
|
| 118 |
+
"grad_norm": 0.3750676214694977,
|
| 119 |
+
"learning_rate": 0.0009358595194085028,
|
| 120 |
+
"loss": 0.0948,
|
| 121 |
+
"step": 400
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.78,
|
| 125 |
+
"grad_norm": 0.33498436212539673,
|
| 126 |
+
"learning_rate": 0.0009312384473197783,
|
| 127 |
+
"loss": 0.077,
|
| 128 |
+
"step": 425
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.82,
|
| 132 |
+
"grad_norm": 0.43420735001564026,
|
| 133 |
+
"learning_rate": 0.0009266173752310536,
|
| 134 |
+
"loss": 0.0729,
|
| 135 |
+
"step": 450
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.87,
|
| 139 |
+
"grad_norm": 1.0590511560440063,
|
| 140 |
+
"learning_rate": 0.0009219963031423291,
|
| 141 |
+
"loss": 0.0816,
|
| 142 |
+
"step": 475
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.92,
|
| 146 |
+
"grad_norm": 0.34223881363868713,
|
| 147 |
+
"learning_rate": 0.0009173752310536044,
|
| 148 |
+
"loss": 0.0567,
|
| 149 |
+
"step": 500
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.96,
|
| 153 |
+
"grad_norm": 0.26913997530937195,
|
| 154 |
+
"learning_rate": 0.0009127541589648799,
|
| 155 |
+
"loss": 0.0681,
|
| 156 |
+
"step": 525
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 1.0,
|
| 160 |
+
"eval_loss": 0.19546650350093842,
|
| 161 |
+
"eval_runtime": 175.4508,
|
| 162 |
+
"eval_samples_per_second": 4.634,
|
| 163 |
+
"eval_steps_per_second": 0.775,
|
| 164 |
+
"step": 546
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 1.01,
|
| 168 |
+
"grad_norm": 0.41807565093040466,
|
| 169 |
+
"learning_rate": 0.0009081330868761552,
|
| 170 |
+
"loss": 0.0365,
|
| 171 |
+
"step": 550
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 1.05,
|
| 175 |
+
"grad_norm": 0.8459017872810364,
|
| 176 |
+
"learning_rate": 0.0009035120147874307,
|
| 177 |
+
"loss": 0.0481,
|
| 178 |
+
"step": 575
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 1.1,
|
| 182 |
+
"grad_norm": 0.21556143462657928,
|
| 183 |
+
"learning_rate": 0.000898890942698706,
|
| 184 |
+
"loss": 0.0426,
|
| 185 |
+
"step": 600
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 1.14,
|
| 189 |
+
"grad_norm": 0.06540340185165405,
|
| 190 |
+
"learning_rate": 0.0008942698706099815,
|
| 191 |
+
"loss": 0.0435,
|
| 192 |
+
"step": 625
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 1.19,
|
| 196 |
+
"grad_norm": 0.8853626251220703,
|
| 197 |
+
"learning_rate": 0.0008896487985212569,
|
| 198 |
+
"loss": 0.0668,
|
| 199 |
+
"step": 650
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 1.24,
|
| 203 |
+
"grad_norm": 0.2227557897567749,
|
| 204 |
+
"learning_rate": 0.0008850277264325323,
|
| 205 |
+
"loss": 0.0373,
|
| 206 |
+
"step": 675
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 1.28,
|
| 210 |
+
"grad_norm": 0.5395527482032776,
|
| 211 |
+
"learning_rate": 0.0008804066543438077,
|
| 212 |
+
"loss": 0.0505,
|
| 213 |
+
"step": 700
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 1.33,
|
| 217 |
+
"grad_norm": 0.47810012102127075,
|
| 218 |
+
"learning_rate": 0.0008757855822550833,
|
| 219 |
+
"loss": 0.0619,
|
| 220 |
+
"step": 725
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.37,
|
| 224 |
+
"grad_norm": 0.41824525594711304,
|
| 225 |
+
"learning_rate": 0.0008711645101663586,
|
| 226 |
+
"loss": 0.0505,
|
| 227 |
+
"step": 750
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 1.42,
|
| 231 |
+
"grad_norm": 0.35845717787742615,
|
| 232 |
+
"learning_rate": 0.0008665434380776341,
|
| 233 |
+
"loss": 0.0556,
|
| 234 |
+
"step": 775
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 1.47,
|
| 238 |
+
"grad_norm": 0.591626763343811,
|
| 239 |
+
"learning_rate": 0.0008619223659889095,
|
| 240 |
+
"loss": 0.0452,
|
| 241 |
+
"step": 800
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 1.51,
|
| 245 |
+
"grad_norm": 0.52753746509552,
|
| 246 |
+
"learning_rate": 0.0008573012939001849,
|
| 247 |
+
"loss": 0.0413,
|
| 248 |
+
"step": 825
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.56,
|
| 252 |
+
"grad_norm": 0.17933356761932373,
|
| 253 |
+
"learning_rate": 0.0008526802218114603,
|
| 254 |
+
"loss": 0.0264,
|
| 255 |
+
"step": 850
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 1.6,
|
| 259 |
+
"grad_norm": 0.4725402891635895,
|
| 260 |
+
"learning_rate": 0.0008480591497227357,
|
| 261 |
+
"loss": 0.0503,
|
| 262 |
+
"step": 875
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.65,
|
| 266 |
+
"grad_norm": 0.43168240785598755,
|
| 267 |
+
"learning_rate": 0.0008434380776340112,
|
| 268 |
+
"loss": 0.0543,
|
| 269 |
+
"step": 900
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.69,
|
| 273 |
+
"grad_norm": 0.15935927629470825,
|
| 274 |
+
"learning_rate": 0.0008388170055452865,
|
| 275 |
+
"loss": 0.0625,
|
| 276 |
+
"step": 925
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 1.74,
|
| 280 |
+
"grad_norm": 0.1527830809354782,
|
| 281 |
+
"learning_rate": 0.000834195933456562,
|
| 282 |
+
"loss": 0.0338,
|
| 283 |
+
"step": 950
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 1.79,
|
| 287 |
+
"grad_norm": 0.6140448451042175,
|
| 288 |
+
"learning_rate": 0.0008295748613678373,
|
| 289 |
+
"loss": 0.0391,
|
| 290 |
+
"step": 975
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 1.83,
|
| 294 |
+
"grad_norm": 0.34989482164382935,
|
| 295 |
+
"learning_rate": 0.0008249537892791128,
|
| 296 |
+
"loss": 0.0428,
|
| 297 |
+
"step": 1000
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 1.88,
|
| 301 |
+
"grad_norm": 1.364334225654602,
|
| 302 |
+
"learning_rate": 0.0008203327171903881,
|
| 303 |
+
"loss": 0.0452,
|
| 304 |
+
"step": 1025
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.92,
|
| 308 |
+
"grad_norm": 0.4410119354724884,
|
| 309 |
+
"learning_rate": 0.0008157116451016636,
|
| 310 |
+
"loss": 0.0519,
|
| 311 |
+
"step": 1050
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 1.97,
|
| 315 |
+
"grad_norm": 0.018156496807932854,
|
| 316 |
+
"learning_rate": 0.000811090573012939,
|
| 317 |
+
"loss": 0.0367,
|
| 318 |
+
"step": 1075
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 2.0,
|
| 322 |
+
"eval_loss": 0.19916145503520966,
|
| 323 |
+
"eval_runtime": 174.1191,
|
| 324 |
+
"eval_samples_per_second": 4.669,
|
| 325 |
+
"eval_steps_per_second": 0.781,
|
| 326 |
+
"step": 1092
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 2.01,
|
| 330 |
+
"grad_norm": 0.2884193956851959,
|
| 331 |
+
"learning_rate": 0.0008064695009242144,
|
| 332 |
+
"loss": 0.0437,
|
| 333 |
+
"step": 1100
|
| 334 |
+
},
|
| 335 |
+
{
|
| 336 |
+
"epoch": 2.06,
|
| 337 |
+
"grad_norm": 0.14955410361289978,
|
| 338 |
+
"learning_rate": 0.0008018484288354898,
|
| 339 |
+
"loss": 0.0332,
|
| 340 |
+
"step": 1125
|
| 341 |
+
},
|
| 342 |
+
{
|
| 343 |
+
"epoch": 2.11,
|
| 344 |
+
"grad_norm": 0.016795417293906212,
|
| 345 |
+
"learning_rate": 0.0007972273567467652,
|
| 346 |
+
"loss": 0.0614,
|
| 347 |
+
"step": 1150
|
| 348 |
+
},
|
| 349 |
+
{
|
| 350 |
+
"epoch": 2.15,
|
| 351 |
+
"grad_norm": 0.272935688495636,
|
| 352 |
+
"learning_rate": 0.0007926062846580406,
|
| 353 |
+
"loss": 0.0354,
|
| 354 |
+
"step": 1175
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.2,
|
| 358 |
+
"grad_norm": 0.181545689702034,
|
| 359 |
+
"learning_rate": 0.0007879852125693162,
|
| 360 |
+
"loss": 0.0477,
|
| 361 |
+
"step": 1200
|
| 362 |
+
},
|
| 363 |
+
{
|
| 364 |
+
"epoch": 2.24,
|
| 365 |
+
"grad_norm": 0.14347945153713226,
|
| 366 |
+
"learning_rate": 0.0007833641404805915,
|
| 367 |
+
"loss": 0.0285,
|
| 368 |
+
"step": 1225
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 2.29,
|
| 372 |
+
"grad_norm": 0.1175965741276741,
|
| 373 |
+
"learning_rate": 0.000778743068391867,
|
| 374 |
+
"loss": 0.0398,
|
| 375 |
+
"step": 1250
|
| 376 |
+
},
|
| 377 |
+
{
|
| 378 |
+
"epoch": 2.34,
|
| 379 |
+
"grad_norm": 0.089854396879673,
|
| 380 |
+
"learning_rate": 0.0007741219963031424,
|
| 381 |
+
"loss": 0.0473,
|
| 382 |
+
"step": 1275
|
| 383 |
+
},
|
| 384 |
+
{
|
| 385 |
+
"epoch": 2.38,
|
| 386 |
+
"grad_norm": 0.3998129665851593,
|
| 387 |
+
"learning_rate": 0.0007695009242144178,
|
| 388 |
+
"loss": 0.0425,
|
| 389 |
+
"step": 1300
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 2.43,
|
| 393 |
+
"grad_norm": 0.42039960622787476,
|
| 394 |
+
"learning_rate": 0.0007648798521256932,
|
| 395 |
+
"loss": 0.0404,
|
| 396 |
+
"step": 1325
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 2.47,
|
| 400 |
+
"grad_norm": 0.3940460979938507,
|
| 401 |
+
"learning_rate": 0.0007602587800369686,
|
| 402 |
+
"loss": 0.0381,
|
| 403 |
+
"step": 1350
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 2.52,
|
| 407 |
+
"grad_norm": 0.37924668192863464,
|
| 408 |
+
"learning_rate": 0.0007556377079482441,
|
| 409 |
+
"loss": 0.0397,
|
| 410 |
+
"step": 1375
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 2.56,
|
| 414 |
+
"grad_norm": 0.46505168080329895,
|
| 415 |
+
"learning_rate": 0.0007510166358595194,
|
| 416 |
+
"loss": 0.0348,
|
| 417 |
+
"step": 1400
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 2.61,
|
| 421 |
+
"grad_norm": 0.2604403495788574,
|
| 422 |
+
"learning_rate": 0.0007463955637707949,
|
| 423 |
+
"loss": 0.036,
|
| 424 |
+
"step": 1425
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 2.66,
|
| 428 |
+
"grad_norm": 0.4742681384086609,
|
| 429 |
+
"learning_rate": 0.0007417744916820702,
|
| 430 |
+
"loss": 0.0216,
|
| 431 |
+
"step": 1450
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 2.7,
|
| 435 |
+
"grad_norm": 0.5116605162620544,
|
| 436 |
+
"learning_rate": 0.0007371534195933457,
|
| 437 |
+
"loss": 0.0281,
|
| 438 |
+
"step": 1475
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 2.75,
|
| 442 |
+
"grad_norm": 0.8539583683013916,
|
| 443 |
+
"learning_rate": 0.000732532347504621,
|
| 444 |
+
"loss": 0.0347,
|
| 445 |
+
"step": 1500
|
| 446 |
+
},
|
| 447 |
+
{
|
| 448 |
+
"epoch": 2.79,
|
| 449 |
+
"grad_norm": 1.8664207458496094,
|
| 450 |
+
"learning_rate": 0.0007279112754158965,
|
| 451 |
+
"loss": 0.045,
|
| 452 |
+
"step": 1525
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 2.84,
|
| 456 |
+
"grad_norm": 0.18790672719478607,
|
| 457 |
+
"learning_rate": 0.0007232902033271719,
|
| 458 |
+
"loss": 0.0233,
|
| 459 |
+
"step": 1550
|
| 460 |
+
},
|
| 461 |
+
{
|
| 462 |
+
"epoch": 2.88,
|
| 463 |
+
"grad_norm": 0.31298670172691345,
|
| 464 |
+
"learning_rate": 0.0007186691312384473,
|
| 465 |
+
"loss": 0.0337,
|
| 466 |
+
"step": 1575
|
| 467 |
+
},
|
| 468 |
+
{
|
| 469 |
+
"epoch": 2.93,
|
| 470 |
+
"grad_norm": 0.8353794813156128,
|
| 471 |
+
"learning_rate": 0.0007140480591497227,
|
| 472 |
+
"loss": 0.0387,
|
| 473 |
+
"step": 1600
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.98,
|
| 477 |
+
"grad_norm": 0.08966954797506332,
|
| 478 |
+
"learning_rate": 0.0007094269870609981,
|
| 479 |
+
"loss": 0.0382,
|
| 480 |
+
"step": 1625
|
| 481 |
+
},
|
| 482 |
+
{
|
| 483 |
+
"epoch": 3.0,
|
| 484 |
+
"eval_loss": 0.18573200702667236,
|
| 485 |
+
"eval_runtime": 175.912,
|
| 486 |
+
"eval_samples_per_second": 4.622,
|
| 487 |
+
"eval_steps_per_second": 0.773,
|
| 488 |
+
"step": 1638
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 3.02,
|
| 492 |
+
"grad_norm": 0.039618588984012604,
|
| 493 |
+
"learning_rate": 0.0007048059149722735,
|
| 494 |
+
"loss": 0.0353,
|
| 495 |
+
"step": 1650
|
| 496 |
+
},
|
| 497 |
+
{
|
| 498 |
+
"epoch": 3.07,
|
| 499 |
+
"grad_norm": 0.3156013488769531,
|
| 500 |
+
"learning_rate": 0.000700184842883549,
|
| 501 |
+
"loss": 0.0318,
|
| 502 |
+
"step": 1675
|
| 503 |
+
},
|
| 504 |
+
{
|
| 505 |
+
"epoch": 3.11,
|
| 506 |
+
"grad_norm": 1.9033042192459106,
|
| 507 |
+
"learning_rate": 0.0006955637707948245,
|
| 508 |
+
"loss": 0.0273,
|
| 509 |
+
"step": 1700
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 3.16,
|
| 513 |
+
"grad_norm": 0.32316115498542786,
|
| 514 |
+
"learning_rate": 0.0006909426987060999,
|
| 515 |
+
"loss": 0.035,
|
| 516 |
+
"step": 1725
|
| 517 |
+
},
|
| 518 |
+
{
|
| 519 |
+
"epoch": 3.21,
|
| 520 |
+
"grad_norm": 0.5656726956367493,
|
| 521 |
+
"learning_rate": 0.0006863216266173753,
|
| 522 |
+
"loss": 0.0311,
|
| 523 |
+
"step": 1750
|
| 524 |
+
},
|
| 525 |
+
{
|
| 526 |
+
"epoch": 3.25,
|
| 527 |
+
"grad_norm": 0.032537270337343216,
|
| 528 |
+
"learning_rate": 0.0006817005545286507,
|
| 529 |
+
"loss": 0.0274,
|
| 530 |
+
"step": 1775
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 3.3,
|
| 534 |
+
"grad_norm": 0.30572062730789185,
|
| 535 |
+
"learning_rate": 0.0006770794824399261,
|
| 536 |
+
"loss": 0.0362,
|
| 537 |
+
"step": 1800
|
| 538 |
+
},
|
| 539 |
+
{
|
| 540 |
+
"epoch": 3.34,
|
| 541 |
+
"grad_norm": 0.3374157249927521,
|
| 542 |
+
"learning_rate": 0.0006724584103512015,
|
| 543 |
+
"loss": 0.0305,
|
| 544 |
+
"step": 1825
|
| 545 |
+
},
|
| 546 |
+
{
|
| 547 |
+
"epoch": 3.39,
|
| 548 |
+
"grad_norm": 0.1089138388633728,
|
| 549 |
+
"learning_rate": 0.000667837338262477,
|
| 550 |
+
"loss": 0.0301,
|
| 551 |
+
"step": 1850
|
| 552 |
+
},
|
| 553 |
+
{
|
| 554 |
+
"epoch": 3.43,
|
| 555 |
+
"grad_norm": 0.10849720984697342,
|
| 556 |
+
"learning_rate": 0.0006632162661737523,
|
| 557 |
+
"loss": 0.0241,
|
| 558 |
+
"step": 1875
|
| 559 |
+
},
|
| 560 |
+
{
|
| 561 |
+
"epoch": 3.48,
|
| 562 |
+
"grad_norm": 0.11349553614854813,
|
| 563 |
+
"learning_rate": 0.0006585951940850278,
|
| 564 |
+
"loss": 0.0218,
|
| 565 |
+
"step": 1900
|
| 566 |
+
},
|
| 567 |
+
{
|
| 568 |
+
"epoch": 3.53,
|
| 569 |
+
"grad_norm": 0.3237963616847992,
|
| 570 |
+
"learning_rate": 0.0006539741219963031,
|
| 571 |
+
"loss": 0.019,
|
| 572 |
+
"step": 1925
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 3.57,
|
| 576 |
+
"grad_norm": 0.5101845860481262,
|
| 577 |
+
"learning_rate": 0.0006493530499075786,
|
| 578 |
+
"loss": 0.0244,
|
| 579 |
+
"step": 1950
|
| 580 |
+
},
|
| 581 |
+
{
|
| 582 |
+
"epoch": 3.62,
|
| 583 |
+
"grad_norm": 0.010137775912880898,
|
| 584 |
+
"learning_rate": 0.0006447319778188539,
|
| 585 |
+
"loss": 0.0304,
|
| 586 |
+
"step": 1975
|
| 587 |
+
},
|
| 588 |
+
{
|
| 589 |
+
"epoch": 3.66,
|
| 590 |
+
"grad_norm": 0.8954480886459351,
|
| 591 |
+
"learning_rate": 0.0006401109057301294,
|
| 592 |
+
"loss": 0.0306,
|
| 593 |
+
"step": 2000
|
| 594 |
+
},
|
| 595 |
+
{
|
| 596 |
+
"epoch": 3.71,
|
| 597 |
+
"grad_norm": 0.014889650978147984,
|
| 598 |
+
"learning_rate": 0.0006354898336414048,
|
| 599 |
+
"loss": 0.0251,
|
| 600 |
+
"step": 2025
|
| 601 |
+
},
|
| 602 |
+
{
|
| 603 |
+
"epoch": 3.75,
|
| 604 |
+
"grad_norm": 0.0878029614686966,
|
| 605 |
+
"learning_rate": 0.0006308687615526802,
|
| 606 |
+
"loss": 0.0341,
|
| 607 |
+
"step": 2050
|
| 608 |
+
},
|
| 609 |
+
{
|
| 610 |
+
"epoch": 3.8,
|
| 611 |
+
"grad_norm": 0.13351218402385712,
|
| 612 |
+
"learning_rate": 0.0006262476894639556,
|
| 613 |
+
"loss": 0.0246,
|
| 614 |
+
"step": 2075
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 3.85,
|
| 618 |
+
"grad_norm": 0.3208947479724884,
|
| 619 |
+
"learning_rate": 0.000621626617375231,
|
| 620 |
+
"loss": 0.0248,
|
| 621 |
+
"step": 2100
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 3.89,
|
| 625 |
+
"grad_norm": 0.42570358514785767,
|
| 626 |
+
"learning_rate": 0.0006170055452865064,
|
| 627 |
+
"loss": 0.0358,
|
| 628 |
+
"step": 2125
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 3.94,
|
| 632 |
+
"grad_norm": 0.131515234708786,
|
| 633 |
+
"learning_rate": 0.000612384473197782,
|
| 634 |
+
"loss": 0.0355,
|
| 635 |
+
"step": 2150
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 3.98,
|
| 639 |
+
"grad_norm": 0.12344180792570114,
|
| 640 |
+
"learning_rate": 0.0006077634011090574,
|
| 641 |
+
"loss": 0.0189,
|
| 642 |
+
"step": 2175
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 4.0,
|
| 646 |
+
"eval_loss": 0.19701677560806274,
|
| 647 |
+
"eval_runtime": 176.7126,
|
| 648 |
+
"eval_samples_per_second": 4.601,
|
| 649 |
+
"eval_steps_per_second": 0.77,
|
| 650 |
+
"step": 2184
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 4.03,
|
| 654 |
+
"grad_norm": 0.10986749082803726,
|
| 655 |
+
"learning_rate": 0.0006031423290203328,
|
| 656 |
+
"loss": 0.0259,
|
| 657 |
+
"step": 2200
|
| 658 |
+
},
|
| 659 |
+
{
|
| 660 |
+
"epoch": 4.08,
|
| 661 |
+
"grad_norm": 0.46528518199920654,
|
| 662 |
+
"learning_rate": 0.0005985212569316082,
|
| 663 |
+
"loss": 0.022,
|
| 664 |
+
"step": 2225
|
| 665 |
+
},
|
| 666 |
+
{
|
| 667 |
+
"epoch": 4.12,
|
| 668 |
+
"grad_norm": 0.2069913148880005,
|
| 669 |
+
"learning_rate": 0.0005939001848428836,
|
| 670 |
+
"loss": 0.015,
|
| 671 |
+
"step": 2250
|
| 672 |
+
},
|
| 673 |
+
{
|
| 674 |
+
"epoch": 4.17,
|
| 675 |
+
"grad_norm": 0.34658578038215637,
|
| 676 |
+
"learning_rate": 0.000589279112754159,
|
| 677 |
+
"loss": 0.0299,
|
| 678 |
+
"step": 2275
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 4.21,
|
| 682 |
+
"grad_norm": 0.18868118524551392,
|
| 683 |
+
"learning_rate": 0.0005846580406654344,
|
| 684 |
+
"loss": 0.0174,
|
| 685 |
+
"step": 2300
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 4.26,
|
| 689 |
+
"grad_norm": 0.33069688081741333,
|
| 690 |
+
"learning_rate": 0.0005800369685767099,
|
| 691 |
+
"loss": 0.0216,
|
| 692 |
+
"step": 2325
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 4.3,
|
| 696 |
+
"grad_norm": 0.7511343955993652,
|
| 697 |
+
"learning_rate": 0.0005754158964879852,
|
| 698 |
+
"loss": 0.0157,
|
| 699 |
+
"step": 2350
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 4.35,
|
| 703 |
+
"grad_norm": 0.27277225255966187,
|
| 704 |
+
"learning_rate": 0.0005707948243992607,
|
| 705 |
+
"loss": 0.0198,
|
| 706 |
+
"step": 2375
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 4.4,
|
| 710 |
+
"grad_norm": 2.3098878860473633,
|
| 711 |
+
"learning_rate": 0.000566173752310536,
|
| 712 |
+
"loss": 0.026,
|
| 713 |
+
"step": 2400
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 4.44,
|
| 717 |
+
"grad_norm": 0.39823707938194275,
|
| 718 |
+
"learning_rate": 0.0005615526802218115,
|
| 719 |
+
"loss": 0.0118,
|
| 720 |
+
"step": 2425
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 4.49,
|
| 724 |
+
"grad_norm": 0.2773701250553131,
|
| 725 |
+
"learning_rate": 0.0005569316081330868,
|
| 726 |
+
"loss": 0.0319,
|
| 727 |
+
"step": 2450
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 4.53,
|
| 731 |
+
"grad_norm": 0.2549929916858673,
|
| 732 |
+
"learning_rate": 0.0005523105360443623,
|
| 733 |
+
"loss": 0.0164,
|
| 734 |
+
"step": 2475
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 4.58,
|
| 738 |
+
"grad_norm": 3.1059272289276123,
|
| 739 |
+
"learning_rate": 0.0005476894639556377,
|
| 740 |
+
"loss": 0.0231,
|
| 741 |
+
"step": 2500
|
| 742 |
+
},
|
| 743 |
+
{
|
| 744 |
+
"epoch": 4.62,
|
| 745 |
+
"grad_norm": 0.10516056418418884,
|
| 746 |
+
"learning_rate": 0.0005430683918669131,
|
| 747 |
+
"loss": 0.0262,
|
| 748 |
+
"step": 2525
|
| 749 |
+
},
|
| 750 |
+
{
|
| 751 |
+
"epoch": 4.67,
|
| 752 |
+
"grad_norm": 0.046087902039289474,
|
| 753 |
+
"learning_rate": 0.0005384473197781885,
|
| 754 |
+
"loss": 0.0212,
|
| 755 |
+
"step": 2550
|
| 756 |
+
},
|
| 757 |
+
{
|
| 758 |
+
"epoch": 4.72,
|
| 759 |
+
"grad_norm": 0.9207663536071777,
|
| 760 |
+
"learning_rate": 0.0005338262476894639,
|
| 761 |
+
"loss": 0.018,
|
| 762 |
+
"step": 2575
|
| 763 |
+
},
|
| 764 |
+
{
|
| 765 |
+
"epoch": 4.76,
|
| 766 |
+
"grad_norm": 0.5687919855117798,
|
| 767 |
+
"learning_rate": 0.0005292051756007393,
|
| 768 |
+
"loss": 0.0255,
|
| 769 |
+
"step": 2600
|
| 770 |
+
},
|
| 771 |
+
{
|
| 772 |
+
"epoch": 4.81,
|
| 773 |
+
"grad_norm": 0.006184098310768604,
|
| 774 |
+
"learning_rate": 0.0005245841035120147,
|
| 775 |
+
"loss": 0.0207,
|
| 776 |
+
"step": 2625
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 4.85,
|
| 780 |
+
"grad_norm": 0.5442487597465515,
|
| 781 |
+
"learning_rate": 0.0005199630314232903,
|
| 782 |
+
"loss": 0.0192,
|
| 783 |
+
"step": 2650
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 4.9,
|
| 787 |
+
"grad_norm": 0.031753990799188614,
|
| 788 |
+
"learning_rate": 0.0005153419593345657,
|
| 789 |
+
"loss": 0.015,
|
| 790 |
+
"step": 2675
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 4.95,
|
| 794 |
+
"grad_norm": 0.022051149979233742,
|
| 795 |
+
"learning_rate": 0.0005107208872458411,
|
| 796 |
+
"loss": 0.0252,
|
| 797 |
+
"step": 2700
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 4.99,
|
| 801 |
+
"grad_norm": 0.17456993460655212,
|
| 802 |
+
"learning_rate": 0.0005060998151571165,
|
| 803 |
+
"loss": 0.0274,
|
| 804 |
+
"step": 2725
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 5.0,
|
| 808 |
+
"eval_loss": 0.18937160074710846,
|
| 809 |
+
"eval_runtime": 177.0599,
|
| 810 |
+
"eval_samples_per_second": 4.592,
|
| 811 |
+
"eval_steps_per_second": 0.768,
|
| 812 |
+
"step": 2730
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 5.04,
|
| 816 |
+
"grad_norm": 0.643435537815094,
|
| 817 |
+
"learning_rate": 0.0005014787430683919,
|
| 818 |
+
"loss": 0.0211,
|
| 819 |
+
"step": 2750
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 5.08,
|
| 823 |
+
"grad_norm": 0.35862746834754944,
|
| 824 |
+
"learning_rate": 0.0004968576709796673,
|
| 825 |
+
"loss": 0.0073,
|
| 826 |
+
"step": 2775
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 5.13,
|
| 830 |
+
"grad_norm": 0.5732066035270691,
|
| 831 |
+
"learning_rate": 0.0004922365988909427,
|
| 832 |
+
"loss": 0.0083,
|
| 833 |
+
"step": 2800
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 5.17,
|
| 837 |
+
"grad_norm": 0.21464449167251587,
|
| 838 |
+
"learning_rate": 0.0004876155268022181,
|
| 839 |
+
"loss": 0.0104,
|
| 840 |
+
"step": 2825
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 5.22,
|
| 844 |
+
"grad_norm": 0.1674581915140152,
|
| 845 |
+
"learning_rate": 0.0004829944547134935,
|
| 846 |
+
"loss": 0.0093,
|
| 847 |
+
"step": 2850
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 5.27,
|
| 851 |
+
"grad_norm": 0.03593946248292923,
|
| 852 |
+
"learning_rate": 0.000478373382624769,
|
| 853 |
+
"loss": 0.0119,
|
| 854 |
+
"step": 2875
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 5.31,
|
| 858 |
+
"grad_norm": 0.18074722588062286,
|
| 859 |
+
"learning_rate": 0.0004737523105360444,
|
| 860 |
+
"loss": 0.0097,
|
| 861 |
+
"step": 2900
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 5.36,
|
| 865 |
+
"grad_norm": 0.06277300417423248,
|
| 866 |
+
"learning_rate": 0.0004691312384473198,
|
| 867 |
+
"loss": 0.0137,
|
| 868 |
+
"step": 2925
|
| 869 |
+
},
|
| 870 |
+
{
|
| 871 |
+
"epoch": 5.4,
|
| 872 |
+
"grad_norm": 0.20016886293888092,
|
| 873 |
+
"learning_rate": 0.0004645101663585952,
|
| 874 |
+
"loss": 0.0204,
|
| 875 |
+
"step": 2950
|
| 876 |
+
},
|
| 877 |
+
{
|
| 878 |
+
"epoch": 5.45,
|
| 879 |
+
"grad_norm": 0.1815144419670105,
|
| 880 |
+
"learning_rate": 0.0004598890942698706,
|
| 881 |
+
"loss": 0.0162,
|
| 882 |
+
"step": 2975
|
| 883 |
+
},
|
| 884 |
+
{
|
| 885 |
+
"epoch": 5.49,
|
| 886 |
+
"grad_norm": 0.5112192034721375,
|
| 887 |
+
"learning_rate": 0.00045526802218114607,
|
| 888 |
+
"loss": 0.0131,
|
| 889 |
+
"step": 3000
|
| 890 |
+
},
|
| 891 |
+
{
|
| 892 |
+
"epoch": 5.54,
|
| 893 |
+
"grad_norm": 0.1796441674232483,
|
| 894 |
+
"learning_rate": 0.0004506469500924215,
|
| 895 |
+
"loss": 0.0176,
|
| 896 |
+
"step": 3025
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 5.59,
|
| 900 |
+
"grad_norm": 0.4108269214630127,
|
| 901 |
+
"learning_rate": 0.0004460258780036969,
|
| 902 |
+
"loss": 0.0196,
|
| 903 |
+
"step": 3050
|
| 904 |
+
},
|
| 905 |
+
{
|
| 906 |
+
"epoch": 5.63,
|
| 907 |
+
"grad_norm": 0.4271663427352905,
|
| 908 |
+
"learning_rate": 0.0004414048059149723,
|
| 909 |
+
"loss": 0.017,
|
| 910 |
+
"step": 3075
|
| 911 |
+
},
|
| 912 |
+
{
|
| 913 |
+
"epoch": 5.68,
|
| 914 |
+
"grad_norm": 0.2981961667537689,
|
| 915 |
+
"learning_rate": 0.0004367837338262477,
|
| 916 |
+
"loss": 0.012,
|
| 917 |
+
"step": 3100
|
| 918 |
+
},
|
| 919 |
+
{
|
| 920 |
+
"epoch": 5.72,
|
| 921 |
+
"grad_norm": 0.392818808555603,
|
| 922 |
+
"learning_rate": 0.0004321626617375231,
|
| 923 |
+
"loss": 0.0115,
|
| 924 |
+
"step": 3125
|
| 925 |
+
},
|
| 926 |
+
{
|
| 927 |
+
"epoch": 5.77,
|
| 928 |
+
"grad_norm": 0.00586000457406044,
|
| 929 |
+
"learning_rate": 0.0004275415896487985,
|
| 930 |
+
"loss": 0.0115,
|
| 931 |
+
"step": 3150
|
| 932 |
+
},
|
| 933 |
+
{
|
| 934 |
+
"epoch": 5.82,
|
| 935 |
+
"grad_norm": 0.2224288433790207,
|
| 936 |
+
"learning_rate": 0.0004229205175600739,
|
| 937 |
+
"loss": 0.0127,
|
| 938 |
+
"step": 3175
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 5.86,
|
| 942 |
+
"grad_norm": 0.28421640396118164,
|
| 943 |
+
"learning_rate": 0.00041829944547134933,
|
| 944 |
+
"loss": 0.0124,
|
| 945 |
+
"step": 3200
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 5.91,
|
| 949 |
+
"grad_norm": 0.3791782557964325,
|
| 950 |
+
"learning_rate": 0.00041367837338262474,
|
| 951 |
+
"loss": 0.016,
|
| 952 |
+
"step": 3225
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 5.95,
|
| 956 |
+
"grad_norm": 0.12688513100147247,
|
| 957 |
+
"learning_rate": 0.0004090573012939002,
|
| 958 |
+
"loss": 0.0212,
|
| 959 |
+
"step": 3250
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 6.0,
|
| 963 |
+
"grad_norm": 0.009004692547023296,
|
| 964 |
+
"learning_rate": 0.0004044362292051756,
|
| 965 |
+
"loss": 0.02,
|
| 966 |
+
"step": 3275
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 6.0,
|
| 970 |
+
"eval_loss": 0.18766650557518005,
|
| 971 |
+
"eval_runtime": 177.5092,
|
| 972 |
+
"eval_samples_per_second": 4.58,
|
| 973 |
+
"eval_steps_per_second": 0.766,
|
| 974 |
+
"step": 3276
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 6.04,
|
| 978 |
+
"grad_norm": 0.014529082924127579,
|
| 979 |
+
"learning_rate": 0.000399815157116451,
|
| 980 |
+
"loss": 0.012,
|
| 981 |
+
"step": 3300
|
| 982 |
+
},
|
| 983 |
+
{
|
| 984 |
+
"epoch": 6.09,
|
| 985 |
+
"grad_norm": 0.16003918647766113,
|
| 986 |
+
"learning_rate": 0.0003951940850277264,
|
| 987 |
+
"loss": 0.0078,
|
| 988 |
+
"step": 3325
|
| 989 |
+
},
|
| 990 |
+
{
|
| 991 |
+
"epoch": 6.14,
|
| 992 |
+
"grad_norm": 0.042826466262340546,
|
| 993 |
+
"learning_rate": 0.0003905730129390019,
|
| 994 |
+
"loss": 0.0116,
|
| 995 |
+
"step": 3350
|
| 996 |
+
},
|
| 997 |
+
{
|
| 998 |
+
"epoch": 6.18,
|
| 999 |
+
"grad_norm": 0.0034067954402416945,
|
| 1000 |
+
"learning_rate": 0.0003859519408502773,
|
| 1001 |
+
"loss": 0.006,
|
| 1002 |
+
"step": 3375
|
| 1003 |
+
},
|
| 1004 |
+
{
|
| 1005 |
+
"epoch": 6.23,
|
| 1006 |
+
"grad_norm": 0.005681981332600117,
|
| 1007 |
+
"learning_rate": 0.0003813308687615527,
|
| 1008 |
+
"loss": 0.0088,
|
| 1009 |
+
"step": 3400
|
| 1010 |
+
},
|
| 1011 |
+
{
|
| 1012 |
+
"epoch": 6.27,
|
| 1013 |
+
"grad_norm": 0.05403963476419449,
|
| 1014 |
+
"learning_rate": 0.0003767097966728281,
|
| 1015 |
+
"loss": 0.0104,
|
| 1016 |
+
"step": 3425
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 6.32,
|
| 1020 |
+
"grad_norm": 0.1421121209859848,
|
| 1021 |
+
"learning_rate": 0.0003720887245841035,
|
| 1022 |
+
"loss": 0.0066,
|
| 1023 |
+
"step": 3450
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 6.36,
|
| 1027 |
+
"grad_norm": 0.02004937455058098,
|
| 1028 |
+
"learning_rate": 0.0003674676524953789,
|
| 1029 |
+
"loss": 0.0075,
|
| 1030 |
+
"step": 3475
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 6.41,
|
| 1034 |
+
"grad_norm": 0.009357332251966,
|
| 1035 |
+
"learning_rate": 0.0003628465804066544,
|
| 1036 |
+
"loss": 0.0065,
|
| 1037 |
+
"step": 3500
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 6.46,
|
| 1041 |
+
"grad_norm": 0.01666351594030857,
|
| 1042 |
+
"learning_rate": 0.0003582255083179298,
|
| 1043 |
+
"loss": 0.006,
|
| 1044 |
+
"step": 3525
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 6.5,
|
| 1048 |
+
"grad_norm": 0.24134355783462524,
|
| 1049 |
+
"learning_rate": 0.0003536044362292052,
|
| 1050 |
+
"loss": 0.0088,
|
| 1051 |
+
"step": 3550
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 6.55,
|
| 1055 |
+
"grad_norm": 0.14924415946006775,
|
| 1056 |
+
"learning_rate": 0.0003489833641404806,
|
| 1057 |
+
"loss": 0.007,
|
| 1058 |
+
"step": 3575
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 6.59,
|
| 1062 |
+
"grad_norm": 0.12202003598213196,
|
| 1063 |
+
"learning_rate": 0.000344362292051756,
|
| 1064 |
+
"loss": 0.0101,
|
| 1065 |
+
"step": 3600
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 6.64,
|
| 1069 |
+
"grad_norm": 0.0060227783396840096,
|
| 1070 |
+
"learning_rate": 0.0003397412199630314,
|
| 1071 |
+
"loss": 0.0117,
|
| 1072 |
+
"step": 3625
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 6.68,
|
| 1076 |
+
"grad_norm": 0.3869228959083557,
|
| 1077 |
+
"learning_rate": 0.0003351201478743068,
|
| 1078 |
+
"loss": 0.0151,
|
| 1079 |
+
"step": 3650
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 6.73,
|
| 1083 |
+
"grad_norm": 0.018938152119517326,
|
| 1084 |
+
"learning_rate": 0.00033049907578558223,
|
| 1085 |
+
"loss": 0.0076,
|
| 1086 |
+
"step": 3675
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 6.78,
|
| 1090 |
+
"grad_norm": 0.018859045580029488,
|
| 1091 |
+
"learning_rate": 0.00032587800369685764,
|
| 1092 |
+
"loss": 0.0085,
|
| 1093 |
+
"step": 3700
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 6.82,
|
| 1097 |
+
"grad_norm": 0.08804900199174881,
|
| 1098 |
+
"learning_rate": 0.0003212569316081331,
|
| 1099 |
+
"loss": 0.0097,
|
| 1100 |
+
"step": 3725
|
| 1101 |
+
},
|
| 1102 |
+
{
|
| 1103 |
+
"epoch": 6.87,
|
| 1104 |
+
"grad_norm": 0.3045863211154938,
|
| 1105 |
+
"learning_rate": 0.0003166358595194085,
|
| 1106 |
+
"loss": 0.0132,
|
| 1107 |
+
"step": 3750
|
| 1108 |
+
},
|
| 1109 |
+
{
|
| 1110 |
+
"epoch": 6.91,
|
| 1111 |
+
"grad_norm": 0.022158470004796982,
|
| 1112 |
+
"learning_rate": 0.0003120147874306839,
|
| 1113 |
+
"loss": 0.0124,
|
| 1114 |
+
"step": 3775
|
| 1115 |
+
},
|
| 1116 |
+
{
|
| 1117 |
+
"epoch": 6.96,
|
| 1118 |
+
"grad_norm": 0.15056921541690826,
|
| 1119 |
+
"learning_rate": 0.0003073937153419594,
|
| 1120 |
+
"loss": 0.0087,
|
| 1121 |
+
"step": 3800
|
| 1122 |
+
},
|
| 1123 |
+
{
|
| 1124 |
+
"epoch": 7.0,
|
| 1125 |
+
"eval_loss": 0.19078923761844635,
|
| 1126 |
+
"eval_runtime": 177.5513,
|
| 1127 |
+
"eval_samples_per_second": 4.579,
|
| 1128 |
+
"eval_steps_per_second": 0.766,
|
| 1129 |
+
"step": 3822
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 7.01,
|
| 1133 |
+
"grad_norm": 0.04336291924118996,
|
| 1134 |
+
"learning_rate": 0.0003027726432532348,
|
| 1135 |
+
"loss": 0.0086,
|
| 1136 |
+
"step": 3825
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 7.05,
|
| 1140 |
+
"grad_norm": 0.0327971875667572,
|
| 1141 |
+
"learning_rate": 0.0002981515711645102,
|
| 1142 |
+
"loss": 0.0084,
|
| 1143 |
+
"step": 3850
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 7.1,
|
| 1147 |
+
"grad_norm": 0.0314444899559021,
|
| 1148 |
+
"learning_rate": 0.0002935304990757856,
|
| 1149 |
+
"loss": 0.0048,
|
| 1150 |
+
"step": 3875
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 7.14,
|
| 1154 |
+
"grad_norm": 0.17276029288768768,
|
| 1155 |
+
"learning_rate": 0.000288909426987061,
|
| 1156 |
+
"loss": 0.007,
|
| 1157 |
+
"step": 3900
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 7.19,
|
| 1161 |
+
"grad_norm": 0.18024314939975739,
|
| 1162 |
+
"learning_rate": 0.0002842883548983364,
|
| 1163 |
+
"loss": 0.0074,
|
| 1164 |
+
"step": 3925
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 7.23,
|
| 1168 |
+
"grad_norm": 0.01734893210232258,
|
| 1169 |
+
"learning_rate": 0.0002796672828096118,
|
| 1170 |
+
"loss": 0.0071,
|
| 1171 |
+
"step": 3950
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 7.28,
|
| 1175 |
+
"grad_norm": 0.01721636950969696,
|
| 1176 |
+
"learning_rate": 0.0002750462107208873,
|
| 1177 |
+
"loss": 0.0123,
|
| 1178 |
+
"step": 3975
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 7.33,
|
| 1182 |
+
"grad_norm": 0.03225923702120781,
|
| 1183 |
+
"learning_rate": 0.0002704251386321627,
|
| 1184 |
+
"loss": 0.0061,
|
| 1185 |
+
"step": 4000
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 7.37,
|
| 1189 |
+
"grad_norm": 0.10785706341266632,
|
| 1190 |
+
"learning_rate": 0.0002658040665434381,
|
| 1191 |
+
"loss": 0.0071,
|
| 1192 |
+
"step": 4025
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 7.42,
|
| 1196 |
+
"grad_norm": 0.02195531316101551,
|
| 1197 |
+
"learning_rate": 0.0002611829944547135,
|
| 1198 |
+
"loss": 0.0067,
|
| 1199 |
+
"step": 4050
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 7.46,
|
| 1203 |
+
"grad_norm": 0.025887854397296906,
|
| 1204 |
+
"learning_rate": 0.0002565619223659889,
|
| 1205 |
+
"loss": 0.0072,
|
| 1206 |
+
"step": 4075
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 7.51,
|
| 1210 |
+
"grad_norm": 1.8573029041290283,
|
| 1211 |
+
"learning_rate": 0.0002519408502772643,
|
| 1212 |
+
"loss": 0.0044,
|
| 1213 |
+
"step": 4100
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 7.55,
|
| 1217 |
+
"grad_norm": 0.41556769609451294,
|
| 1218 |
+
"learning_rate": 0.0002473197781885397,
|
| 1219 |
+
"loss": 0.0076,
|
| 1220 |
+
"step": 4125
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 7.6,
|
| 1224 |
+
"grad_norm": 0.0036406666040420532,
|
| 1225 |
+
"learning_rate": 0.0002426987060998152,
|
| 1226 |
+
"loss": 0.0054,
|
| 1227 |
+
"step": 4150
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 7.65,
|
| 1231 |
+
"grad_norm": 0.1950559765100479,
|
| 1232 |
+
"learning_rate": 0.0002380776340110906,
|
| 1233 |
+
"loss": 0.0052,
|
| 1234 |
+
"step": 4175
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 7.69,
|
| 1238 |
+
"grad_norm": 0.01785474270582199,
|
| 1239 |
+
"learning_rate": 0.000233456561922366,
|
| 1240 |
+
"loss": 0.007,
|
| 1241 |
+
"step": 4200
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 7.74,
|
| 1245 |
+
"grad_norm": 0.26933544874191284,
|
| 1246 |
+
"learning_rate": 0.0002288354898336414,
|
| 1247 |
+
"loss": 0.0048,
|
| 1248 |
+
"step": 4225
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 7.78,
|
| 1252 |
+
"grad_norm": 0.19295917451381683,
|
| 1253 |
+
"learning_rate": 0.00022421441774491682,
|
| 1254 |
+
"loss": 0.0035,
|
| 1255 |
+
"step": 4250
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 7.83,
|
| 1259 |
+
"grad_norm": 0.008535887114703655,
|
| 1260 |
+
"learning_rate": 0.00021959334565619225,
|
| 1261 |
+
"loss": 0.0063,
|
| 1262 |
+
"step": 4275
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 7.88,
|
| 1266 |
+
"grad_norm": 0.16601914167404175,
|
| 1267 |
+
"learning_rate": 0.00021497227356746766,
|
| 1268 |
+
"loss": 0.0049,
|
| 1269 |
+
"step": 4300
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 7.92,
|
| 1273 |
+
"grad_norm": 0.25450438261032104,
|
| 1274 |
+
"learning_rate": 0.00021035120147874306,
|
| 1275 |
+
"loss": 0.0069,
|
| 1276 |
+
"step": 4325
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 7.97,
|
| 1280 |
+
"grad_norm": 0.049375709146261215,
|
| 1281 |
+
"learning_rate": 0.00020573012939001847,
|
| 1282 |
+
"loss": 0.0066,
|
| 1283 |
+
"step": 4350
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 8.0,
|
| 1287 |
+
"eval_loss": 0.2085200548171997,
|
| 1288 |
+
"eval_runtime": 177.9722,
|
| 1289 |
+
"eval_samples_per_second": 4.568,
|
| 1290 |
+
"eval_steps_per_second": 0.764,
|
| 1291 |
+
"step": 4368
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 8.01,
|
| 1295 |
+
"grad_norm": 0.06922808289527893,
|
| 1296 |
+
"learning_rate": 0.00020110905730129388,
|
| 1297 |
+
"loss": 0.0042,
|
| 1298 |
+
"step": 4375
|
| 1299 |
+
},
|
| 1300 |
+
{
|
| 1301 |
+
"epoch": 8.06,
|
| 1302 |
+
"grad_norm": 0.04170389473438263,
|
| 1303 |
+
"learning_rate": 0.00019648798521256934,
|
| 1304 |
+
"loss": 0.0037,
|
| 1305 |
+
"step": 4400
|
| 1306 |
+
},
|
| 1307 |
+
{
|
| 1308 |
+
"epoch": 8.1,
|
| 1309 |
+
"grad_norm": 0.010052547790110111,
|
| 1310 |
+
"learning_rate": 0.00019186691312384475,
|
| 1311 |
+
"loss": 0.0029,
|
| 1312 |
+
"step": 4425
|
| 1313 |
+
},
|
| 1314 |
+
{
|
| 1315 |
+
"epoch": 8.15,
|
| 1316 |
+
"grad_norm": 0.25184884667396545,
|
| 1317 |
+
"learning_rate": 0.00018724584103512016,
|
| 1318 |
+
"loss": 0.0039,
|
| 1319 |
+
"step": 4450
|
| 1320 |
+
},
|
| 1321 |
+
{
|
| 1322 |
+
"epoch": 8.2,
|
| 1323 |
+
"grad_norm": 0.07106045633554459,
|
| 1324 |
+
"learning_rate": 0.00018262476894639556,
|
| 1325 |
+
"loss": 0.0039,
|
| 1326 |
+
"step": 4475
|
| 1327 |
+
},
|
| 1328 |
+
{
|
| 1329 |
+
"epoch": 8.24,
|
| 1330 |
+
"grad_norm": 0.002000249456614256,
|
| 1331 |
+
"learning_rate": 0.00017800369685767097,
|
| 1332 |
+
"loss": 0.0056,
|
| 1333 |
+
"step": 4500
|
| 1334 |
+
},
|
| 1335 |
+
{
|
| 1336 |
+
"epoch": 8.29,
|
| 1337 |
+
"grad_norm": 0.025201383978128433,
|
| 1338 |
+
"learning_rate": 0.0001733826247689464,
|
| 1339 |
+
"loss": 0.0031,
|
| 1340 |
+
"step": 4525
|
| 1341 |
+
},
|
| 1342 |
+
{
|
| 1343 |
+
"epoch": 8.33,
|
| 1344 |
+
"grad_norm": 0.0007307173800654709,
|
| 1345 |
+
"learning_rate": 0.0001687615526802218,
|
| 1346 |
+
"loss": 0.0031,
|
| 1347 |
+
"step": 4550
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 8.38,
|
| 1351 |
+
"grad_norm": 0.010259617120027542,
|
| 1352 |
+
"learning_rate": 0.00016414048059149722,
|
| 1353 |
+
"loss": 0.0036,
|
| 1354 |
+
"step": 4575
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 8.42,
|
| 1358 |
+
"grad_norm": 0.004237270914018154,
|
| 1359 |
+
"learning_rate": 0.00015951940850277263,
|
| 1360 |
+
"loss": 0.0038,
|
| 1361 |
+
"step": 4600
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 8.47,
|
| 1365 |
+
"grad_norm": 0.02443511225283146,
|
| 1366 |
+
"learning_rate": 0.0001548983364140481,
|
| 1367 |
+
"loss": 0.0044,
|
| 1368 |
+
"step": 4625
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 8.52,
|
| 1372 |
+
"grad_norm": 0.039590246975421906,
|
| 1373 |
+
"learning_rate": 0.0001502772643253235,
|
| 1374 |
+
"loss": 0.0019,
|
| 1375 |
+
"step": 4650
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 8.56,
|
| 1379 |
+
"grad_norm": 0.30276018381118774,
|
| 1380 |
+
"learning_rate": 0.0001456561922365989,
|
| 1381 |
+
"loss": 0.0023,
|
| 1382 |
+
"step": 4675
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 8.61,
|
| 1386 |
+
"grad_norm": 0.05218060687184334,
|
| 1387 |
+
"learning_rate": 0.0001410351201478743,
|
| 1388 |
+
"loss": 0.003,
|
| 1389 |
+
"step": 4700
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 8.65,
|
| 1393 |
+
"grad_norm": 0.02608703263103962,
|
| 1394 |
+
"learning_rate": 0.00013641404805914972,
|
| 1395 |
+
"loss": 0.0027,
|
| 1396 |
+
"step": 4725
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 8.7,
|
| 1400 |
+
"grad_norm": 0.007796884514391422,
|
| 1401 |
+
"learning_rate": 0.00013179297597042515,
|
| 1402 |
+
"loss": 0.0038,
|
| 1403 |
+
"step": 4750
|
| 1404 |
+
},
|
| 1405 |
+
{
|
| 1406 |
+
"epoch": 8.75,
|
| 1407 |
+
"grad_norm": 0.008572472259402275,
|
| 1408 |
+
"learning_rate": 0.00012717190388170056,
|
| 1409 |
+
"loss": 0.0018,
|
| 1410 |
+
"step": 4775
|
| 1411 |
+
},
|
| 1412 |
+
{
|
| 1413 |
+
"epoch": 8.79,
|
| 1414 |
+
"grad_norm": 0.0034019711893051863,
|
| 1415 |
+
"learning_rate": 0.00012255083179297597,
|
| 1416 |
+
"loss": 0.003,
|
| 1417 |
+
"step": 4800
|
| 1418 |
+
},
|
| 1419 |
+
{
|
| 1420 |
+
"epoch": 8.84,
|
| 1421 |
+
"grad_norm": 0.003986136056482792,
|
| 1422 |
+
"learning_rate": 0.00011792975970425139,
|
| 1423 |
+
"loss": 0.0047,
|
| 1424 |
+
"step": 4825
|
| 1425 |
+
},
|
| 1426 |
+
{
|
| 1427 |
+
"epoch": 8.88,
|
| 1428 |
+
"grad_norm": 0.055789873003959656,
|
| 1429 |
+
"learning_rate": 0.00011330868761552681,
|
| 1430 |
+
"loss": 0.0021,
|
| 1431 |
+
"step": 4850
|
| 1432 |
+
},
|
| 1433 |
+
{
|
| 1434 |
+
"epoch": 8.93,
|
| 1435 |
+
"grad_norm": 0.07775359600782394,
|
| 1436 |
+
"learning_rate": 0.00010868761552680221,
|
| 1437 |
+
"loss": 0.0032,
|
| 1438 |
+
"step": 4875
|
| 1439 |
+
},
|
| 1440 |
+
{
|
| 1441 |
+
"epoch": 8.97,
|
| 1442 |
+
"grad_norm": 0.0017645555781200528,
|
| 1443 |
+
"learning_rate": 0.00010406654343807764,
|
| 1444 |
+
"loss": 0.0055,
|
| 1445 |
+
"step": 4900
|
| 1446 |
+
},
|
| 1447 |
+
{
|
| 1448 |
+
"epoch": 9.0,
|
| 1449 |
+
"eval_loss": 0.21004897356033325,
|
| 1450 |
+
"eval_runtime": 178.8955,
|
| 1451 |
+
"eval_samples_per_second": 4.545,
|
| 1452 |
+
"eval_steps_per_second": 0.76,
|
| 1453 |
+
"step": 4914
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 9.02,
|
| 1457 |
+
"grad_norm": 0.22125497460365295,
|
| 1458 |
+
"learning_rate": 9.944547134935306e-05,
|
| 1459 |
+
"loss": 0.0031,
|
| 1460 |
+
"step": 4925
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 9.07,
|
| 1464 |
+
"grad_norm": 0.003768475726246834,
|
| 1465 |
+
"learning_rate": 9.482439926062846e-05,
|
| 1466 |
+
"loss": 0.0013,
|
| 1467 |
+
"step": 4950
|
| 1468 |
+
},
|
| 1469 |
+
{
|
| 1470 |
+
"epoch": 9.11,
|
| 1471 |
+
"grad_norm": 0.013520549982786179,
|
| 1472 |
+
"learning_rate": 9.020332717190388e-05,
|
| 1473 |
+
"loss": 0.0025,
|
| 1474 |
+
"step": 4975
|
| 1475 |
+
},
|
| 1476 |
+
{
|
| 1477 |
+
"epoch": 9.16,
|
| 1478 |
+
"grad_norm": 0.009503871202468872,
|
| 1479 |
+
"learning_rate": 8.558225508317929e-05,
|
| 1480 |
+
"loss": 0.0024,
|
| 1481 |
+
"step": 5000
|
| 1482 |
+
},
|
| 1483 |
+
{
|
| 1484 |
+
"epoch": 9.2,
|
| 1485 |
+
"grad_norm": 0.0057460549287498,
|
| 1486 |
+
"learning_rate": 8.096118299445473e-05,
|
| 1487 |
+
"loss": 0.0015,
|
| 1488 |
+
"step": 5025
|
| 1489 |
+
},
|
| 1490 |
+
{
|
| 1491 |
+
"epoch": 9.25,
|
| 1492 |
+
"grad_norm": 0.06969017535448074,
|
| 1493 |
+
"learning_rate": 7.634011090573013e-05,
|
| 1494 |
+
"loss": 0.0017,
|
| 1495 |
+
"step": 5050
|
| 1496 |
+
},
|
| 1497 |
+
{
|
| 1498 |
+
"epoch": 9.29,
|
| 1499 |
+
"grad_norm": 0.1530989110469818,
|
| 1500 |
+
"learning_rate": 7.171903881700554e-05,
|
| 1501 |
+
"loss": 0.0022,
|
| 1502 |
+
"step": 5075
|
| 1503 |
+
},
|
| 1504 |
+
{
|
| 1505 |
+
"epoch": 9.34,
|
| 1506 |
+
"grad_norm": 0.1752089112997055,
|
| 1507 |
+
"learning_rate": 6.709796672828096e-05,
|
| 1508 |
+
"loss": 0.0018,
|
| 1509 |
+
"step": 5100
|
| 1510 |
+
},
|
| 1511 |
+
{
|
| 1512 |
+
"epoch": 9.39,
|
| 1513 |
+
"grad_norm": 0.023138588294386864,
|
| 1514 |
+
"learning_rate": 6.247689463955638e-05,
|
| 1515 |
+
"loss": 0.0014,
|
| 1516 |
+
"step": 5125
|
| 1517 |
+
},
|
| 1518 |
+
{
|
| 1519 |
+
"epoch": 9.43,
|
| 1520 |
+
"grad_norm": 0.005098209250718355,
|
| 1521 |
+
"learning_rate": 5.785582255083179e-05,
|
| 1522 |
+
"loss": 0.0012,
|
| 1523 |
+
"step": 5150
|
| 1524 |
+
},
|
| 1525 |
+
{
|
| 1526 |
+
"epoch": 9.48,
|
| 1527 |
+
"grad_norm": 0.007919879630208015,
|
| 1528 |
+
"learning_rate": 5.323475046210721e-05,
|
| 1529 |
+
"loss": 0.0023,
|
| 1530 |
+
"step": 5175
|
| 1531 |
+
},
|
| 1532 |
+
{
|
| 1533 |
+
"epoch": 9.52,
|
| 1534 |
+
"grad_norm": 0.0019298276165500283,
|
| 1535 |
+
"learning_rate": 4.8613678373382625e-05,
|
| 1536 |
+
"loss": 0.0015,
|
| 1537 |
+
"step": 5200
|
| 1538 |
+
},
|
| 1539 |
+
{
|
| 1540 |
+
"epoch": 9.57,
|
| 1541 |
+
"grad_norm": 0.0023822402581572533,
|
| 1542 |
+
"learning_rate": 4.3992606284658045e-05,
|
| 1543 |
+
"loss": 0.0011,
|
| 1544 |
+
"step": 5225
|
| 1545 |
+
},
|
| 1546 |
+
{
|
| 1547 |
+
"epoch": 9.62,
|
| 1548 |
+
"grad_norm": 0.03612617775797844,
|
| 1549 |
+
"learning_rate": 3.937153419593346e-05,
|
| 1550 |
+
"loss": 0.001,
|
| 1551 |
+
"step": 5250
|
| 1552 |
+
},
|
| 1553 |
+
{
|
| 1554 |
+
"epoch": 9.66,
|
| 1555 |
+
"grad_norm": 0.03683371841907501,
|
| 1556 |
+
"learning_rate": 3.4750462107208874e-05,
|
| 1557 |
+
"loss": 0.0016,
|
| 1558 |
+
"step": 5275
|
| 1559 |
+
},
|
| 1560 |
+
{
|
| 1561 |
+
"epoch": 9.71,
|
| 1562 |
+
"grad_norm": 0.04906224459409714,
|
| 1563 |
+
"learning_rate": 3.012939001848429e-05,
|
| 1564 |
+
"loss": 0.0022,
|
| 1565 |
+
"step": 5300
|
| 1566 |
+
},
|
| 1567 |
+
{
|
| 1568 |
+
"epoch": 9.75,
|
| 1569 |
+
"grad_norm": 0.08069704473018646,
|
| 1570 |
+
"learning_rate": 2.5508317929759705e-05,
|
| 1571 |
+
"loss": 0.0015,
|
| 1572 |
+
"step": 5325
|
| 1573 |
+
},
|
| 1574 |
+
{
|
| 1575 |
+
"epoch": 9.8,
|
| 1576 |
+
"grad_norm": 0.13353778421878815,
|
| 1577 |
+
"learning_rate": 2.088724584103512e-05,
|
| 1578 |
+
"loss": 0.0013,
|
| 1579 |
+
"step": 5350
|
| 1580 |
+
},
|
| 1581 |
+
{
|
| 1582 |
+
"epoch": 9.84,
|
| 1583 |
+
"grad_norm": 0.10152421146631241,
|
| 1584 |
+
"learning_rate": 1.6266173752310537e-05,
|
| 1585 |
+
"loss": 0.0015,
|
| 1586 |
+
"step": 5375
|
| 1587 |
+
},
|
| 1588 |
+
{
|
| 1589 |
+
"epoch": 9.89,
|
| 1590 |
+
"grad_norm": 0.010886043310165405,
|
| 1591 |
+
"learning_rate": 1.1645101663585952e-05,
|
| 1592 |
+
"loss": 0.0017,
|
| 1593 |
+
"step": 5400
|
| 1594 |
+
},
|
| 1595 |
+
{
|
| 1596 |
+
"epoch": 9.94,
|
| 1597 |
+
"grad_norm": 0.009057571180164814,
|
| 1598 |
+
"learning_rate": 7.024029574861368e-06,
|
| 1599 |
+
"loss": 0.0016,
|
| 1600 |
+
"step": 5425
|
| 1601 |
+
},
|
| 1602 |
+
{
|
| 1603 |
+
"epoch": 9.98,
|
| 1604 |
+
"grad_norm": 0.020738158375024796,
|
| 1605 |
+
"learning_rate": 2.402957486136784e-06,
|
| 1606 |
+
"loss": 0.0013,
|
| 1607 |
+
"step": 5450
|
| 1608 |
+
},
|
| 1609 |
+
{
|
| 1610 |
+
"epoch": 10.0,
|
| 1611 |
+
"eval_loss": 0.21373072266578674,
|
| 1612 |
+
"eval_runtime": 177.6097,
|
| 1613 |
+
"eval_samples_per_second": 4.577,
|
| 1614 |
+
"eval_steps_per_second": 0.766,
|
| 1615 |
+
"step": 5460
|
| 1616 |
+
},
|
| 1617 |
+
{
|
| 1618 |
+
"epoch": 10.0,
|
| 1619 |
+
"step": 5460,
|
| 1620 |
+
"total_flos": 9.7789895073792e+18,
|
| 1621 |
+
"train_loss": 0.05130936206342318,
|
| 1622 |
+
"train_runtime": 10728.949,
|
| 1623 |
+
"train_samples_per_second": 3.051,
|
| 1624 |
+
"train_steps_per_second": 0.509
|
| 1625 |
+
}
|
| 1626 |
+
],
|
| 1627 |
+
"logging_steps": 25,
|
| 1628 |
+
"max_steps": 5460,
|
| 1629 |
+
"num_input_tokens_seen": 0,
|
| 1630 |
+
"num_train_epochs": 10,
|
| 1631 |
+
"save_steps": 500,
|
| 1632 |
+
"total_flos": 9.7789895073792e+18,
|
| 1633 |
+
"train_batch_size": 6,
|
| 1634 |
+
"trial_name": null,
|
| 1635 |
+
"trial_params": null
|
| 1636 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:3d450ec50cab4188af1d2f839c282d646159565242704493acf7a0046664a3f1
|
| 3 |
+
size 5112
|