End of training
Browse files- README.md +89 -0
- adapter_config.json +32 -0
- adapter_model.safetensors +3 -0
- trainer_state.json +1317 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,89 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
library_name: peft
|
| 3 |
+
license: other
|
| 4 |
+
base_model: deepseek-ai/deepseek-coder-1.3b-base
|
| 5 |
+
tags:
|
| 6 |
+
- generated_from_trainer
|
| 7 |
+
model-index:
|
| 8 |
+
- name: lemexp-afp-small-thms-deepseek-coder-1.3b-base
|
| 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 |
+
# lemexp-afp-small-thms-deepseek-coder-1.3b-base
|
| 16 |
+
|
| 17 |
+
This model is a fine-tuned version of [deepseek-ai/deepseek-coder-1.3b-base](https://huggingface.co/deepseek-ai/deepseek-coder-1.3b-base) on an unknown dataset.
|
| 18 |
+
It achieves the following results on the evaluation set:
|
| 19 |
+
- Loss: 0.1166
|
| 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: 0.0002
|
| 39 |
+
- train_batch_size: 2
|
| 40 |
+
- eval_batch_size: 2
|
| 41 |
+
- seed: 42
|
| 42 |
+
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
|
| 43 |
+
- lr_scheduler_type: linear
|
| 44 |
+
- lr_scheduler_warmup_steps: 100
|
| 45 |
+
- num_epochs: 6
|
| 46 |
+
- mixed_precision_training: Native AMP
|
| 47 |
+
|
| 48 |
+
### Training results
|
| 49 |
+
|
| 50 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 51 |
+
|:-------------:|:------:|:-----:|:---------------:|
|
| 52 |
+
| 0.4601 | 0.2001 | 2492 | 0.3758 |
|
| 53 |
+
| 0.369 | 0.4001 | 4984 | 0.3327 |
|
| 54 |
+
| 0.327 | 0.6002 | 7476 | 0.2894 |
|
| 55 |
+
| 0.2988 | 0.8003 | 9968 | 0.2848 |
|
| 56 |
+
| 0.2873 | 1.0003 | 12460 | 0.2698 |
|
| 57 |
+
| 0.2487 | 1.2004 | 14952 | 0.2669 |
|
| 58 |
+
| 0.2371 | 1.4004 | 17444 | 0.2430 |
|
| 59 |
+
| 0.2324 | 1.6005 | 19936 | 0.2373 |
|
| 60 |
+
| 0.2278 | 1.8006 | 22428 | 0.2372 |
|
| 61 |
+
| 0.2212 | 2.0006 | 24920 | 0.2190 |
|
| 62 |
+
| 0.1991 | 2.2007 | 27412 | 0.2131 |
|
| 63 |
+
| 0.1854 | 2.4008 | 29904 | 0.2092 |
|
| 64 |
+
| 0.1769 | 2.6008 | 32396 | 0.2034 |
|
| 65 |
+
| 0.1818 | 2.8009 | 34888 | 0.1904 |
|
| 66 |
+
| 0.1798 | 3.0010 | 37380 | 0.1980 |
|
| 67 |
+
| 0.1656 | 3.2010 | 39872 | 0.1820 |
|
| 68 |
+
| 0.1533 | 3.4011 | 42364 | 0.1724 |
|
| 69 |
+
| 0.1557 | 3.6012 | 44856 | 0.1687 |
|
| 70 |
+
| 0.1536 | 3.8012 | 47348 | 0.1721 |
|
| 71 |
+
| 0.1531 | 4.0013 | 49840 | 0.1634 |
|
| 72 |
+
| 0.1256 | 4.2013 | 52332 | 0.1529 |
|
| 73 |
+
| 0.1274 | 4.4014 | 54824 | 0.1471 |
|
| 74 |
+
| 0.1294 | 4.6015 | 57316 | 0.1433 |
|
| 75 |
+
| 0.1209 | 4.8015 | 59808 | 0.1375 |
|
| 76 |
+
| 0.1298 | 5.0016 | 62300 | 0.1308 |
|
| 77 |
+
| 0.1064 | 5.2017 | 64792 | 0.1269 |
|
| 78 |
+
| 0.1063 | 5.4017 | 67284 | 0.1262 |
|
| 79 |
+
| 0.1036 | 5.6018 | 69776 | 0.1213 |
|
| 80 |
+
| 0.1084 | 5.8019 | 72268 | 0.1166 |
|
| 81 |
+
|
| 82 |
+
|
| 83 |
+
### Framework versions
|
| 84 |
+
|
| 85 |
+
- PEFT 0.14.0
|
| 86 |
+
- Transformers 4.47.0
|
| 87 |
+
- Pytorch 2.5.1+cu124
|
| 88 |
+
- Datasets 3.2.0
|
| 89 |
+
- Tokenizers 0.21.0
|
adapter_config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"base_model_name_or_path": "deepseek-ai/deepseek-coder-1.3b-base",
|
| 5 |
+
"bias": "none",
|
| 6 |
+
"eva_config": null,
|
| 7 |
+
"exclude_modules": null,
|
| 8 |
+
"fan_in_fan_out": false,
|
| 9 |
+
"inference_mode": true,
|
| 10 |
+
"init_lora_weights": true,
|
| 11 |
+
"layer_replication": null,
|
| 12 |
+
"layers_pattern": null,
|
| 13 |
+
"layers_to_transform": null,
|
| 14 |
+
"loftq_config": {},
|
| 15 |
+
"lora_alpha": 32,
|
| 16 |
+
"lora_bias": false,
|
| 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": 8,
|
| 23 |
+
"rank_pattern": {},
|
| 24 |
+
"revision": null,
|
| 25 |
+
"target_modules": [
|
| 26 |
+
"q_proj",
|
| 27 |
+
"v_proj"
|
| 28 |
+
],
|
| 29 |
+
"task_type": "CAUSAL_LM",
|
| 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 sha256:4c326065c3f57e18c68a637feff9e7a5ba6a4ff17dc5a50a47b316fd1556a0db
|
| 3 |
+
size 531035104
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1317 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 6.0,
|
| 5 |
+
"eval_steps": 2492,
|
| 6 |
+
"global_step": 74736,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.040141297366730895,
|
| 13 |
+
"grad_norm": 1.3413853645324707,
|
| 14 |
+
"learning_rate": 0.00019893081086874966,
|
| 15 |
+
"loss": 0.8055,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.08028259473346179,
|
| 20 |
+
"grad_norm": 1.8721694946289062,
|
| 21 |
+
"learning_rate": 0.00019759365453668473,
|
| 22 |
+
"loss": 0.5632,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.12042389210019268,
|
| 27 |
+
"grad_norm": 1.127410650253296,
|
| 28 |
+
"learning_rate": 0.0001962538185326116,
|
| 29 |
+
"loss": 0.5087,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.16056518946692358,
|
| 34 |
+
"grad_norm": 1.7598766088485718,
|
| 35 |
+
"learning_rate": 0.00019491398252853852,
|
| 36 |
+
"loss": 0.4601,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.20006422607578678,
|
| 41 |
+
"eval_loss": 0.37580204010009766,
|
| 42 |
+
"eval_runtime": 9.2262,
|
| 43 |
+
"eval_samples_per_second": 10.839,
|
| 44 |
+
"eval_steps_per_second": 5.419,
|
| 45 |
+
"step": 2492
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.20070648683365447,
|
| 49 |
+
"grad_norm": 1.5294134616851807,
|
| 50 |
+
"learning_rate": 0.00019357414652446543,
|
| 51 |
+
"loss": 0.4423,
|
| 52 |
+
"step": 2500
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.24084778420038536,
|
| 56 |
+
"grad_norm": 2.4775636196136475,
|
| 57 |
+
"learning_rate": 0.00019223699019240047,
|
| 58 |
+
"loss": 0.4201,
|
| 59 |
+
"step": 3000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.2809890815671163,
|
| 63 |
+
"grad_norm": 1.5586802959442139,
|
| 64 |
+
"learning_rate": 0.00019089715418832737,
|
| 65 |
+
"loss": 0.3942,
|
| 66 |
+
"step": 3500
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.32113037893384716,
|
| 70 |
+
"grad_norm": 2.8762762546539307,
|
| 71 |
+
"learning_rate": 0.00018955731818425425,
|
| 72 |
+
"loss": 0.3846,
|
| 73 |
+
"step": 4000
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.36127167630057805,
|
| 77 |
+
"grad_norm": 1.1510844230651855,
|
| 78 |
+
"learning_rate": 0.00018821748218018116,
|
| 79 |
+
"loss": 0.369,
|
| 80 |
+
"step": 4500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.40012845215157355,
|
| 84 |
+
"eval_loss": 0.3327239751815796,
|
| 85 |
+
"eval_runtime": 9.2684,
|
| 86 |
+
"eval_samples_per_second": 10.789,
|
| 87 |
+
"eval_steps_per_second": 5.395,
|
| 88 |
+
"step": 4984
|
| 89 |
+
},
|
| 90 |
+
{
|
| 91 |
+
"epoch": 0.40141297366730894,
|
| 92 |
+
"grad_norm": 2.390737771987915,
|
| 93 |
+
"learning_rate": 0.00018687764617610804,
|
| 94 |
+
"loss": 0.3662,
|
| 95 |
+
"step": 5000
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.4415542710340398,
|
| 99 |
+
"grad_norm": 1.9558954238891602,
|
| 100 |
+
"learning_rate": 0.00018553781017203495,
|
| 101 |
+
"loss": 0.3516,
|
| 102 |
+
"step": 5500
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.4816955684007707,
|
| 106 |
+
"grad_norm": 1.7609800100326538,
|
| 107 |
+
"learning_rate": 0.00018419797416796186,
|
| 108 |
+
"loss": 0.3399,
|
| 109 |
+
"step": 6000
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.5218368657675017,
|
| 113 |
+
"grad_norm": 2.026273488998413,
|
| 114 |
+
"learning_rate": 0.00018285813816388874,
|
| 115 |
+
"loss": 0.3374,
|
| 116 |
+
"step": 6500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.5619781631342325,
|
| 120 |
+
"grad_norm": 1.3401412963867188,
|
| 121 |
+
"learning_rate": 0.0001815209818318238,
|
| 122 |
+
"loss": 0.327,
|
| 123 |
+
"step": 7000
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.6001926782273603,
|
| 127 |
+
"eval_loss": 0.2893865406513214,
|
| 128 |
+
"eval_runtime": 9.2199,
|
| 129 |
+
"eval_samples_per_second": 10.846,
|
| 130 |
+
"eval_steps_per_second": 5.423,
|
| 131 |
+
"step": 7476
|
| 132 |
+
},
|
| 133 |
+
{
|
| 134 |
+
"epoch": 0.6021194605009634,
|
| 135 |
+
"grad_norm": 2.0142829418182373,
|
| 136 |
+
"learning_rate": 0.0001801811458277507,
|
| 137 |
+
"loss": 0.3291,
|
| 138 |
+
"step": 7500
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.6422607578676943,
|
| 142 |
+
"grad_norm": 0.9912647604942322,
|
| 143 |
+
"learning_rate": 0.0001788413098236776,
|
| 144 |
+
"loss": 0.3121,
|
| 145 |
+
"step": 8000
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.6824020552344252,
|
| 149 |
+
"grad_norm": 2.552272319793701,
|
| 150 |
+
"learning_rate": 0.00017750415349161264,
|
| 151 |
+
"loss": 0.3117,
|
| 152 |
+
"step": 8500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.7225433526011561,
|
| 156 |
+
"grad_norm": 2.693617582321167,
|
| 157 |
+
"learning_rate": 0.00017616431748753954,
|
| 158 |
+
"loss": 0.3069,
|
| 159 |
+
"step": 9000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.762684649967887,
|
| 163 |
+
"grad_norm": 3.3755452632904053,
|
| 164 |
+
"learning_rate": 0.00017482448148346645,
|
| 165 |
+
"loss": 0.2988,
|
| 166 |
+
"step": 9500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.8002569043031471,
|
| 170 |
+
"eval_loss": 0.28478091955184937,
|
| 171 |
+
"eval_runtime": 9.2844,
|
| 172 |
+
"eval_samples_per_second": 10.771,
|
| 173 |
+
"eval_steps_per_second": 5.385,
|
| 174 |
+
"step": 9968
|
| 175 |
+
},
|
| 176 |
+
{
|
| 177 |
+
"epoch": 0.8028259473346179,
|
| 178 |
+
"grad_norm": 1.348958134651184,
|
| 179 |
+
"learning_rate": 0.00017348464547939333,
|
| 180 |
+
"loss": 0.297,
|
| 181 |
+
"step": 10000
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.8429672447013488,
|
| 185 |
+
"grad_norm": 1.8039051294326782,
|
| 186 |
+
"learning_rate": 0.00017214480947532024,
|
| 187 |
+
"loss": 0.3044,
|
| 188 |
+
"step": 10500
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.8831085420680796,
|
| 192 |
+
"grad_norm": 1.7400553226470947,
|
| 193 |
+
"learning_rate": 0.00017080497347124712,
|
| 194 |
+
"loss": 0.2938,
|
| 195 |
+
"step": 11000
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.9232498394348105,
|
| 199 |
+
"grad_norm": 1.4909226894378662,
|
| 200 |
+
"learning_rate": 0.00016946513746717403,
|
| 201 |
+
"loss": 0.2785,
|
| 202 |
+
"step": 11500
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.9633911368015414,
|
| 206 |
+
"grad_norm": 2.5195188522338867,
|
| 207 |
+
"learning_rate": 0.00016812530146310094,
|
| 208 |
+
"loss": 0.2873,
|
| 209 |
+
"step": 12000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 1.0003211303789339,
|
| 213 |
+
"eval_loss": 0.2697654068470001,
|
| 214 |
+
"eval_runtime": 9.4165,
|
| 215 |
+
"eval_samples_per_second": 10.62,
|
| 216 |
+
"eval_steps_per_second": 5.31,
|
| 217 |
+
"step": 12460
|
| 218 |
+
},
|
| 219 |
+
{
|
| 220 |
+
"epoch": 1.0035324341682723,
|
| 221 |
+
"grad_norm": 2.530623435974121,
|
| 222 |
+
"learning_rate": 0.00016678546545902782,
|
| 223 |
+
"loss": 0.2723,
|
| 224 |
+
"step": 12500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 1.0436737315350033,
|
| 228 |
+
"grad_norm": 1.2821416854858398,
|
| 229 |
+
"learning_rate": 0.00016544830912696288,
|
| 230 |
+
"loss": 0.2632,
|
| 231 |
+
"step": 13000
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 1.083815028901734,
|
| 235 |
+
"grad_norm": 1.7591134309768677,
|
| 236 |
+
"learning_rate": 0.00016410847312288976,
|
| 237 |
+
"loss": 0.2633,
|
| 238 |
+
"step": 13500
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 1.123956326268465,
|
| 242 |
+
"grad_norm": 1.4008091688156128,
|
| 243 |
+
"learning_rate": 0.00016276863711881667,
|
| 244 |
+
"loss": 0.245,
|
| 245 |
+
"step": 14000
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 1.1640976236351959,
|
| 249 |
+
"grad_norm": 1.567612886428833,
|
| 250 |
+
"learning_rate": 0.00016142880111474355,
|
| 251 |
+
"loss": 0.2487,
|
| 252 |
+
"step": 14500
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 1.2003853564547207,
|
| 256 |
+
"eval_loss": 0.26689666509628296,
|
| 257 |
+
"eval_runtime": 9.3095,
|
| 258 |
+
"eval_samples_per_second": 10.742,
|
| 259 |
+
"eval_steps_per_second": 5.371,
|
| 260 |
+
"step": 14952
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 1.2042389210019269,
|
| 264 |
+
"grad_norm": 1.9705049991607666,
|
| 265 |
+
"learning_rate": 0.00016008896511067046,
|
| 266 |
+
"loss": 0.247,
|
| 267 |
+
"step": 15000
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 1.2443802183686576,
|
| 271 |
+
"grad_norm": 1.31686270236969,
|
| 272 |
+
"learning_rate": 0.00015875180877860553,
|
| 273 |
+
"loss": 0.2459,
|
| 274 |
+
"step": 15500
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 1.2845215157353886,
|
| 278 |
+
"grad_norm": 1.3225802183151245,
|
| 279 |
+
"learning_rate": 0.0001574119727745324,
|
| 280 |
+
"loss": 0.253,
|
| 281 |
+
"step": 16000
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 1.3246628131021194,
|
| 285 |
+
"grad_norm": 1.7328706979751587,
|
| 286 |
+
"learning_rate": 0.00015607213677045932,
|
| 287 |
+
"loss": 0.2404,
|
| 288 |
+
"step": 16500
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 1.3648041104688504,
|
| 292 |
+
"grad_norm": 1.6794183254241943,
|
| 293 |
+
"learning_rate": 0.00015473498043839433,
|
| 294 |
+
"loss": 0.2371,
|
| 295 |
+
"step": 17000
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 1.4004495825305074,
|
| 299 |
+
"eval_loss": 0.24297775328159332,
|
| 300 |
+
"eval_runtime": 9.4719,
|
| 301 |
+
"eval_samples_per_second": 10.558,
|
| 302 |
+
"eval_steps_per_second": 5.279,
|
| 303 |
+
"step": 17444
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 1.4049454078355812,
|
| 307 |
+
"grad_norm": 1.2241907119750977,
|
| 308 |
+
"learning_rate": 0.00015339514443432124,
|
| 309 |
+
"loss": 0.2407,
|
| 310 |
+
"step": 17500
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 1.4450867052023122,
|
| 314 |
+
"grad_norm": 2.555837631225586,
|
| 315 |
+
"learning_rate": 0.00015205530843024815,
|
| 316 |
+
"loss": 0.2344,
|
| 317 |
+
"step": 18000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 1.485228002569043,
|
| 321 |
+
"grad_norm": 2.1652560234069824,
|
| 322 |
+
"learning_rate": 0.00015071547242617505,
|
| 323 |
+
"loss": 0.2404,
|
| 324 |
+
"step": 18500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.525369299935774,
|
| 328 |
+
"grad_norm": 3.3938980102539062,
|
| 329 |
+
"learning_rate": 0.00014937563642210196,
|
| 330 |
+
"loss": 0.2383,
|
| 331 |
+
"step": 19000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.565510597302505,
|
| 335 |
+
"grad_norm": 2.078359365463257,
|
| 336 |
+
"learning_rate": 0.00014803580041802884,
|
| 337 |
+
"loss": 0.2324,
|
| 338 |
+
"step": 19500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.6005138086062942,
|
| 342 |
+
"eval_loss": 0.23732736706733704,
|
| 343 |
+
"eval_runtime": 9.3828,
|
| 344 |
+
"eval_samples_per_second": 10.658,
|
| 345 |
+
"eval_steps_per_second": 5.329,
|
| 346 |
+
"step": 19936
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 1.6056518946692357,
|
| 350 |
+
"grad_norm": 2.344526767730713,
|
| 351 |
+
"learning_rate": 0.00014669596441395575,
|
| 352 |
+
"loss": 0.2337,
|
| 353 |
+
"step": 20000
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.6457931920359665,
|
| 357 |
+
"grad_norm": 3.3463504314422607,
|
| 358 |
+
"learning_rate": 0.00014535612840988263,
|
| 359 |
+
"loss": 0.2269,
|
| 360 |
+
"step": 20500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.6859344894026975,
|
| 364 |
+
"grad_norm": 1.8101106882095337,
|
| 365 |
+
"learning_rate": 0.00014401629240580954,
|
| 366 |
+
"loss": 0.2281,
|
| 367 |
+
"step": 21000
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.7260757867694285,
|
| 371 |
+
"grad_norm": 2.2214066982269287,
|
| 372 |
+
"learning_rate": 0.00014267645640173645,
|
| 373 |
+
"loss": 0.2299,
|
| 374 |
+
"step": 21500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.7662170841361593,
|
| 378 |
+
"grad_norm": 3.5231621265411377,
|
| 379 |
+
"learning_rate": 0.00014133930006967149,
|
| 380 |
+
"loss": 0.2278,
|
| 381 |
+
"step": 22000
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.800578034682081,
|
| 385 |
+
"eval_loss": 0.23720289766788483,
|
| 386 |
+
"eval_runtime": 9.3853,
|
| 387 |
+
"eval_samples_per_second": 10.655,
|
| 388 |
+
"eval_steps_per_second": 5.327,
|
| 389 |
+
"step": 22428
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.80635838150289,
|
| 393 |
+
"grad_norm": 1.503125548362732,
|
| 394 |
+
"learning_rate": 0.0001399994640655984,
|
| 395 |
+
"loss": 0.2286,
|
| 396 |
+
"step": 22500
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 1.846499678869621,
|
| 400 |
+
"grad_norm": 1.3709886074066162,
|
| 401 |
+
"learning_rate": 0.00013865962806152527,
|
| 402 |
+
"loss": 0.2204,
|
| 403 |
+
"step": 23000
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 1.886640976236352,
|
| 407 |
+
"grad_norm": 0.8027909398078918,
|
| 408 |
+
"learning_rate": 0.00013731979205745218,
|
| 409 |
+
"loss": 0.2122,
|
| 410 |
+
"step": 23500
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 1.9267822736030829,
|
| 414 |
+
"grad_norm": 1.0423333644866943,
|
| 415 |
+
"learning_rate": 0.0001359826357253872,
|
| 416 |
+
"loss": 0.2209,
|
| 417 |
+
"step": 24000
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 1.9669235709698136,
|
| 421 |
+
"grad_norm": 3.140822172164917,
|
| 422 |
+
"learning_rate": 0.00013464547939332226,
|
| 423 |
+
"loss": 0.2212,
|
| 424 |
+
"step": 24500
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 2.0006422607578678,
|
| 428 |
+
"eval_loss": 0.21895286440849304,
|
| 429 |
+
"eval_runtime": 9.4416,
|
| 430 |
+
"eval_samples_per_second": 10.591,
|
| 431 |
+
"eval_steps_per_second": 5.296,
|
| 432 |
+
"step": 24920
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 2.0070648683365446,
|
| 436 |
+
"grad_norm": 2.0828654766082764,
|
| 437 |
+
"learning_rate": 0.00013330564338924917,
|
| 438 |
+
"loss": 0.2157,
|
| 439 |
+
"step": 25000
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 2.0472061657032756,
|
| 443 |
+
"grad_norm": 1.9575061798095703,
|
| 444 |
+
"learning_rate": 0.00013196580738517605,
|
| 445 |
+
"loss": 0.1942,
|
| 446 |
+
"step": 25500
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 2.0873474630700066,
|
| 450 |
+
"grad_norm": 2.263435125350952,
|
| 451 |
+
"learning_rate": 0.00013062597138110296,
|
| 452 |
+
"loss": 0.1939,
|
| 453 |
+
"step": 26000
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 2.127488760436737,
|
| 457 |
+
"grad_norm": 1.9007372856140137,
|
| 458 |
+
"learning_rate": 0.00012928613537702984,
|
| 459 |
+
"loss": 0.1964,
|
| 460 |
+
"step": 26500
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 2.167630057803468,
|
| 464 |
+
"grad_norm": 1.8549082279205322,
|
| 465 |
+
"learning_rate": 0.00012794629937295675,
|
| 466 |
+
"loss": 0.1991,
|
| 467 |
+
"step": 27000
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 2.2007064868336546,
|
| 471 |
+
"eval_loss": 0.21308277547359467,
|
| 472 |
+
"eval_runtime": 9.4114,
|
| 473 |
+
"eval_samples_per_second": 10.625,
|
| 474 |
+
"eval_steps_per_second": 5.313,
|
| 475 |
+
"step": 27412
|
| 476 |
+
},
|
| 477 |
+
{
|
| 478 |
+
"epoch": 2.207771355170199,
|
| 479 |
+
"grad_norm": 1.3462762832641602,
|
| 480 |
+
"learning_rate": 0.00012660646336888366,
|
| 481 |
+
"loss": 0.1923,
|
| 482 |
+
"step": 27500
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 2.24791265253693,
|
| 486 |
+
"grad_norm": 1.3775655031204224,
|
| 487 |
+
"learning_rate": 0.00012526662736481056,
|
| 488 |
+
"loss": 0.1956,
|
| 489 |
+
"step": 28000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 2.2880539499036607,
|
| 493 |
+
"grad_norm": 3.3470311164855957,
|
| 494 |
+
"learning_rate": 0.0001239294710327456,
|
| 495 |
+
"loss": 0.188,
|
| 496 |
+
"step": 28500
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 2.3281952472703917,
|
| 500 |
+
"grad_norm": 2.1436495780944824,
|
| 501 |
+
"learning_rate": 0.00012259231470068064,
|
| 502 |
+
"loss": 0.1963,
|
| 503 |
+
"step": 29000
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 2.3683365446371227,
|
| 507 |
+
"grad_norm": 1.8383592367172241,
|
| 508 |
+
"learning_rate": 0.00012125247869660754,
|
| 509 |
+
"loss": 0.1854,
|
| 510 |
+
"step": 29500
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 2.4007707129094413,
|
| 514 |
+
"eval_loss": 0.20915324985980988,
|
| 515 |
+
"eval_runtime": 9.2953,
|
| 516 |
+
"eval_samples_per_second": 10.758,
|
| 517 |
+
"eval_steps_per_second": 5.379,
|
| 518 |
+
"step": 29904
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 2.4084778420038537,
|
| 522 |
+
"grad_norm": 1.6092569828033447,
|
| 523 |
+
"learning_rate": 0.00011991264269253443,
|
| 524 |
+
"loss": 0.1949,
|
| 525 |
+
"step": 30000
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 2.4486191393705843,
|
| 529 |
+
"grad_norm": 1.8148771524429321,
|
| 530 |
+
"learning_rate": 0.00011857280668846134,
|
| 531 |
+
"loss": 0.1902,
|
| 532 |
+
"step": 30500
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 2.4887604367373153,
|
| 536 |
+
"grad_norm": 2.0251662731170654,
|
| 537 |
+
"learning_rate": 0.00011723297068438823,
|
| 538 |
+
"loss": 0.1921,
|
| 539 |
+
"step": 31000
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 2.5289017341040463,
|
| 543 |
+
"grad_norm": 4.4525861740112305,
|
| 544 |
+
"learning_rate": 0.00011589313468031513,
|
| 545 |
+
"loss": 0.1972,
|
| 546 |
+
"step": 31500
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 2.5690430314707773,
|
| 550 |
+
"grad_norm": 2.327648878097534,
|
| 551 |
+
"learning_rate": 0.00011455329867624202,
|
| 552 |
+
"loss": 0.1769,
|
| 553 |
+
"step": 32000
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 2.600834938985228,
|
| 557 |
+
"eval_loss": 0.20336776971817017,
|
| 558 |
+
"eval_runtime": 9.3362,
|
| 559 |
+
"eval_samples_per_second": 10.711,
|
| 560 |
+
"eval_steps_per_second": 5.356,
|
| 561 |
+
"step": 32396
|
| 562 |
+
},
|
| 563 |
+
{
|
| 564 |
+
"epoch": 2.6091843288375083,
|
| 565 |
+
"grad_norm": 1.7865198850631714,
|
| 566 |
+
"learning_rate": 0.00011321346267216893,
|
| 567 |
+
"loss": 0.1888,
|
| 568 |
+
"step": 32500
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 2.649325626204239,
|
| 572 |
+
"grad_norm": 2.9860942363739014,
|
| 573 |
+
"learning_rate": 0.00011187362666809582,
|
| 574 |
+
"loss": 0.1948,
|
| 575 |
+
"step": 33000
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 2.68946692357097,
|
| 579 |
+
"grad_norm": 1.5154398679733276,
|
| 580 |
+
"learning_rate": 0.00011053647033603088,
|
| 581 |
+
"loss": 0.1936,
|
| 582 |
+
"step": 33500
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 2.729608220937701,
|
| 586 |
+
"grad_norm": 1.3628119230270386,
|
| 587 |
+
"learning_rate": 0.00010919663433195777,
|
| 588 |
+
"loss": 0.1826,
|
| 589 |
+
"step": 34000
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 2.7697495183044314,
|
| 593 |
+
"grad_norm": 1.1184334754943848,
|
| 594 |
+
"learning_rate": 0.00010785679832788467,
|
| 595 |
+
"loss": 0.1818,
|
| 596 |
+
"step": 34500
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 2.800899165061015,
|
| 600 |
+
"eval_loss": 0.19036497175693512,
|
| 601 |
+
"eval_runtime": 9.2882,
|
| 602 |
+
"eval_samples_per_second": 10.766,
|
| 603 |
+
"eval_steps_per_second": 5.383,
|
| 604 |
+
"step": 34888
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 2.8098908156711624,
|
| 608 |
+
"grad_norm": 3.3239927291870117,
|
| 609 |
+
"learning_rate": 0.00010651696232381156,
|
| 610 |
+
"loss": 0.1738,
|
| 611 |
+
"step": 35000
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 2.8500321130378934,
|
| 615 |
+
"grad_norm": 1.8970750570297241,
|
| 616 |
+
"learning_rate": 0.00010517712631973846,
|
| 617 |
+
"loss": 0.1848,
|
| 618 |
+
"step": 35500
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 2.8901734104046244,
|
| 622 |
+
"grad_norm": 2.5936954021453857,
|
| 623 |
+
"learning_rate": 0.00010383996998767351,
|
| 624 |
+
"loss": 0.1834,
|
| 625 |
+
"step": 36000
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 2.9303147077713554,
|
| 629 |
+
"grad_norm": 4.450264930725098,
|
| 630 |
+
"learning_rate": 0.00010250013398360042,
|
| 631 |
+
"loss": 0.1768,
|
| 632 |
+
"step": 36500
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 2.970456005138086,
|
| 636 |
+
"grad_norm": 2.1718015670776367,
|
| 637 |
+
"learning_rate": 0.00010116029797952731,
|
| 638 |
+
"loss": 0.1798,
|
| 639 |
+
"step": 37000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 3.0009633911368017,
|
| 643 |
+
"eval_loss": 0.19798487424850464,
|
| 644 |
+
"eval_runtime": 9.4331,
|
| 645 |
+
"eval_samples_per_second": 10.601,
|
| 646 |
+
"eval_steps_per_second": 5.3,
|
| 647 |
+
"step": 37380
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 3.010597302504817,
|
| 651 |
+
"grad_norm": 1.2895421981811523,
|
| 652 |
+
"learning_rate": 9.98204619754542e-05,
|
| 653 |
+
"loss": 0.1679,
|
| 654 |
+
"step": 37500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 3.050738599871548,
|
| 658 |
+
"grad_norm": 2.5621845722198486,
|
| 659 |
+
"learning_rate": 9.848330564338926e-05,
|
| 660 |
+
"loss": 0.1556,
|
| 661 |
+
"step": 38000
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 3.090879897238279,
|
| 665 |
+
"grad_norm": 4.237947940826416,
|
| 666 |
+
"learning_rate": 9.714346963931615e-05,
|
| 667 |
+
"loss": 0.1627,
|
| 668 |
+
"step": 38500
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 3.1310211946050095,
|
| 672 |
+
"grad_norm": 0.6110877990722656,
|
| 673 |
+
"learning_rate": 9.580363363524305e-05,
|
| 674 |
+
"loss": 0.1577,
|
| 675 |
+
"step": 39000
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 3.1711624919717405,
|
| 679 |
+
"grad_norm": 1.6049071550369263,
|
| 680 |
+
"learning_rate": 9.446379763116994e-05,
|
| 681 |
+
"loss": 0.1656,
|
| 682 |
+
"step": 39500
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 3.2010276172125884,
|
| 686 |
+
"eval_loss": 0.18195098638534546,
|
| 687 |
+
"eval_runtime": 9.445,
|
| 688 |
+
"eval_samples_per_second": 10.588,
|
| 689 |
+
"eval_steps_per_second": 5.294,
|
| 690 |
+
"step": 39872
|
| 691 |
+
},
|
| 692 |
+
{
|
| 693 |
+
"epoch": 3.2113037893384715,
|
| 694 |
+
"grad_norm": 1.347450852394104,
|
| 695 |
+
"learning_rate": 9.3126641299105e-05,
|
| 696 |
+
"loss": 0.1523,
|
| 697 |
+
"step": 40000
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 3.2514450867052025,
|
| 701 |
+
"grad_norm": 2.5243570804595947,
|
| 702 |
+
"learning_rate": 9.17868052950319e-05,
|
| 703 |
+
"loss": 0.1549,
|
| 704 |
+
"step": 40500
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 3.291586384071933,
|
| 708 |
+
"grad_norm": 1.4282008409500122,
|
| 709 |
+
"learning_rate": 9.044964896296694e-05,
|
| 710 |
+
"loss": 0.1624,
|
| 711 |
+
"step": 41000
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 3.331727681438664,
|
| 715 |
+
"grad_norm": 1.0618197917938232,
|
| 716 |
+
"learning_rate": 8.910981295889384e-05,
|
| 717 |
+
"loss": 0.1506,
|
| 718 |
+
"step": 41500
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 3.371868978805395,
|
| 722 |
+
"grad_norm": 2.2008793354034424,
|
| 723 |
+
"learning_rate": 8.776997695482073e-05,
|
| 724 |
+
"loss": 0.1533,
|
| 725 |
+
"step": 42000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 3.401091843288375,
|
| 729 |
+
"eval_loss": 0.1724245399236679,
|
| 730 |
+
"eval_runtime": 9.5166,
|
| 731 |
+
"eval_samples_per_second": 10.508,
|
| 732 |
+
"eval_steps_per_second": 5.254,
|
| 733 |
+
"step": 42364
|
| 734 |
+
},
|
| 735 |
+
{
|
| 736 |
+
"epoch": 3.412010276172126,
|
| 737 |
+
"grad_norm": 2.9689488410949707,
|
| 738 |
+
"learning_rate": 8.643014095074763e-05,
|
| 739 |
+
"loss": 0.1587,
|
| 740 |
+
"step": 42500
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 3.4521515735388566,
|
| 744 |
+
"grad_norm": 2.200533628463745,
|
| 745 |
+
"learning_rate": 8.509030494667452e-05,
|
| 746 |
+
"loss": 0.1521,
|
| 747 |
+
"step": 43000
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 3.4922928709055876,
|
| 751 |
+
"grad_norm": 2.433441162109375,
|
| 752 |
+
"learning_rate": 8.375046894260143e-05,
|
| 753 |
+
"loss": 0.1559,
|
| 754 |
+
"step": 43500
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 3.5324341682723186,
|
| 758 |
+
"grad_norm": 1.3238328695297241,
|
| 759 |
+
"learning_rate": 8.241063293852834e-05,
|
| 760 |
+
"loss": 0.1513,
|
| 761 |
+
"step": 44000
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 3.5725754656390496,
|
| 765 |
+
"grad_norm": 1.2658567428588867,
|
| 766 |
+
"learning_rate": 8.107079693445523e-05,
|
| 767 |
+
"loss": 0.1557,
|
| 768 |
+
"step": 44500
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 3.601156069364162,
|
| 772 |
+
"eval_loss": 0.1686776727437973,
|
| 773 |
+
"eval_runtime": 9.5131,
|
| 774 |
+
"eval_samples_per_second": 10.512,
|
| 775 |
+
"eval_steps_per_second": 5.256,
|
| 776 |
+
"step": 44856
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 3.61271676300578,
|
| 780 |
+
"grad_norm": 0.9541091918945312,
|
| 781 |
+
"learning_rate": 7.973096093038212e-05,
|
| 782 |
+
"loss": 0.1538,
|
| 783 |
+
"step": 45000
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 3.652858060372511,
|
| 787 |
+
"grad_norm": 2.3008804321289062,
|
| 788 |
+
"learning_rate": 7.839112492630902e-05,
|
| 789 |
+
"loss": 0.1595,
|
| 790 |
+
"step": 45500
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 3.692999357739242,
|
| 794 |
+
"grad_norm": 0.9546318054199219,
|
| 795 |
+
"learning_rate": 7.705128892223593e-05,
|
| 796 |
+
"loss": 0.1511,
|
| 797 |
+
"step": 46000
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 3.733140655105973,
|
| 801 |
+
"grad_norm": 2.0182559490203857,
|
| 802 |
+
"learning_rate": 7.571413259017097e-05,
|
| 803 |
+
"loss": 0.1524,
|
| 804 |
+
"step": 46500
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 3.773281952472704,
|
| 808 |
+
"grad_norm": 3.313910484313965,
|
| 809 |
+
"learning_rate": 7.437429658609787e-05,
|
| 810 |
+
"loss": 0.1536,
|
| 811 |
+
"step": 47000
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 3.8012202954399488,
|
| 815 |
+
"eval_loss": 0.17214125394821167,
|
| 816 |
+
"eval_runtime": 9.4988,
|
| 817 |
+
"eval_samples_per_second": 10.528,
|
| 818 |
+
"eval_steps_per_second": 5.264,
|
| 819 |
+
"step": 47348
|
| 820 |
+
},
|
| 821 |
+
{
|
| 822 |
+
"epoch": 3.8134232498394347,
|
| 823 |
+
"grad_norm": 1.0791243314743042,
|
| 824 |
+
"learning_rate": 7.303446058202477e-05,
|
| 825 |
+
"loss": 0.1559,
|
| 826 |
+
"step": 47500
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 3.8535645472061657,
|
| 830 |
+
"grad_norm": 1.7102487087249756,
|
| 831 |
+
"learning_rate": 7.169462457795166e-05,
|
| 832 |
+
"loss": 0.1558,
|
| 833 |
+
"step": 48000
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 3.8937058445728967,
|
| 837 |
+
"grad_norm": 1.6259522438049316,
|
| 838 |
+
"learning_rate": 7.03574682458867e-05,
|
| 839 |
+
"loss": 0.1519,
|
| 840 |
+
"step": 48500
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 3.9338471419396273,
|
| 844 |
+
"grad_norm": 1.94523286819458,
|
| 845 |
+
"learning_rate": 6.90176322418136e-05,
|
| 846 |
+
"loss": 0.1498,
|
| 847 |
+
"step": 49000
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 3.9739884393063583,
|
| 851 |
+
"grad_norm": 1.2504680156707764,
|
| 852 |
+
"learning_rate": 6.768047590974865e-05,
|
| 853 |
+
"loss": 0.1531,
|
| 854 |
+
"step": 49500
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 4.0012845215157355,
|
| 858 |
+
"eval_loss": 0.1633564978837967,
|
| 859 |
+
"eval_runtime": 9.4116,
|
| 860 |
+
"eval_samples_per_second": 10.625,
|
| 861 |
+
"eval_steps_per_second": 5.313,
|
| 862 |
+
"step": 49840
|
| 863 |
+
},
|
| 864 |
+
{
|
| 865 |
+
"epoch": 4.014129736673089,
|
| 866 |
+
"grad_norm": 2.535567283630371,
|
| 867 |
+
"learning_rate": 6.634063990567555e-05,
|
| 868 |
+
"loss": 0.1421,
|
| 869 |
+
"step": 50000
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 4.05427103403982,
|
| 873 |
+
"grad_norm": 2.8016157150268555,
|
| 874 |
+
"learning_rate": 6.500348357361059e-05,
|
| 875 |
+
"loss": 0.1304,
|
| 876 |
+
"step": 50500
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 4.094412331406551,
|
| 880 |
+
"grad_norm": 3.923179864883423,
|
| 881 |
+
"learning_rate": 6.366364756953749e-05,
|
| 882 |
+
"loss": 0.1222,
|
| 883 |
+
"step": 51000
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 4.134553628773282,
|
| 887 |
+
"grad_norm": 4.409369945526123,
|
| 888 |
+
"learning_rate": 6.232381156546439e-05,
|
| 889 |
+
"loss": 0.1344,
|
| 890 |
+
"step": 51500
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 4.174694926140013,
|
| 894 |
+
"grad_norm": 1.9566134214401245,
|
| 895 |
+
"learning_rate": 6.0983975561391295e-05,
|
| 896 |
+
"loss": 0.1256,
|
| 897 |
+
"step": 52000
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 4.201348747591522,
|
| 901 |
+
"eval_loss": 0.15292324125766754,
|
| 902 |
+
"eval_runtime": 9.4583,
|
| 903 |
+
"eval_samples_per_second": 10.573,
|
| 904 |
+
"eval_steps_per_second": 5.286,
|
| 905 |
+
"step": 52332
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 4.214836223506744,
|
| 909 |
+
"grad_norm": 1.6278679370880127,
|
| 910 |
+
"learning_rate": 5.964413955731819e-05,
|
| 911 |
+
"loss": 0.1274,
|
| 912 |
+
"step": 52500
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 4.254977520873474,
|
| 916 |
+
"grad_norm": 0.310170441865921,
|
| 917 |
+
"learning_rate": 5.830430355324509e-05,
|
| 918 |
+
"loss": 0.1277,
|
| 919 |
+
"step": 53000
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 4.295118818240206,
|
| 923 |
+
"grad_norm": 2.6923305988311768,
|
| 924 |
+
"learning_rate": 5.6964467549171985e-05,
|
| 925 |
+
"loss": 0.1284,
|
| 926 |
+
"step": 53500
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 4.335260115606936,
|
| 930 |
+
"grad_norm": 2.4311087131500244,
|
| 931 |
+
"learning_rate": 5.562463154509888e-05,
|
| 932 |
+
"loss": 0.1274,
|
| 933 |
+
"step": 54000
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 4.375401412973667,
|
| 937 |
+
"grad_norm": 0.18654285371303558,
|
| 938 |
+
"learning_rate": 5.4287475213033926e-05,
|
| 939 |
+
"loss": 0.1274,
|
| 940 |
+
"step": 54500
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 4.401412973667309,
|
| 944 |
+
"eval_loss": 0.14706894755363464,
|
| 945 |
+
"eval_runtime": 9.5362,
|
| 946 |
+
"eval_samples_per_second": 10.486,
|
| 947 |
+
"eval_steps_per_second": 5.243,
|
| 948 |
+
"step": 54824
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 4.415542710340398,
|
| 952 |
+
"grad_norm": 4.156493663787842,
|
| 953 |
+
"learning_rate": 5.2947639208960834e-05,
|
| 954 |
+
"loss": 0.1257,
|
| 955 |
+
"step": 55000
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 4.455684007707129,
|
| 959 |
+
"grad_norm": 1.4045820236206055,
|
| 960 |
+
"learning_rate": 5.160780320488773e-05,
|
| 961 |
+
"loss": 0.1302,
|
| 962 |
+
"step": 55500
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 4.49582530507386,
|
| 966 |
+
"grad_norm": 0.5079956650733948,
|
| 967 |
+
"learning_rate": 5.026796720081462e-05,
|
| 968 |
+
"loss": 0.1302,
|
| 969 |
+
"step": 56000
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 4.535966602440591,
|
| 973 |
+
"grad_norm": 2.671663522720337,
|
| 974 |
+
"learning_rate": 4.892813119674152e-05,
|
| 975 |
+
"loss": 0.1319,
|
| 976 |
+
"step": 56500
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 4.5761078998073215,
|
| 980 |
+
"grad_norm": 2.7671585083007812,
|
| 981 |
+
"learning_rate": 4.758829519266842e-05,
|
| 982 |
+
"loss": 0.1294,
|
| 983 |
+
"step": 57000
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 4.601477199743096,
|
| 987 |
+
"eval_loss": 0.1432746946811676,
|
| 988 |
+
"eval_runtime": 9.2082,
|
| 989 |
+
"eval_samples_per_second": 10.86,
|
| 990 |
+
"eval_steps_per_second": 5.43,
|
| 991 |
+
"step": 57316
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 4.616249197174053,
|
| 995 |
+
"grad_norm": 4.464232921600342,
|
| 996 |
+
"learning_rate": 4.624845918859532e-05,
|
| 997 |
+
"loss": 0.1236,
|
| 998 |
+
"step": 57500
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 4.6563904945407835,
|
| 1002 |
+
"grad_norm": 0.22100146114826202,
|
| 1003 |
+
"learning_rate": 4.490862318452222e-05,
|
| 1004 |
+
"loss": 0.1214,
|
| 1005 |
+
"step": 58000
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 4.696531791907514,
|
| 1009 |
+
"grad_norm": 3.2810275554656982,
|
| 1010 |
+
"learning_rate": 4.357146685245726e-05,
|
| 1011 |
+
"loss": 0.1299,
|
| 1012 |
+
"step": 58500
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 4.7366730892742455,
|
| 1016 |
+
"grad_norm": 4.431360721588135,
|
| 1017 |
+
"learning_rate": 4.2234310520392307e-05,
|
| 1018 |
+
"loss": 0.129,
|
| 1019 |
+
"step": 59000
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 4.776814386640976,
|
| 1023 |
+
"grad_norm": 2.974076509475708,
|
| 1024 |
+
"learning_rate": 4.08944745163192e-05,
|
| 1025 |
+
"loss": 0.1209,
|
| 1026 |
+
"step": 59500
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 4.801541425818883,
|
| 1030 |
+
"eval_loss": 0.13750909268856049,
|
| 1031 |
+
"eval_runtime": 9.2555,
|
| 1032 |
+
"eval_samples_per_second": 10.804,
|
| 1033 |
+
"eval_steps_per_second": 5.402,
|
| 1034 |
+
"step": 59808
|
| 1035 |
+
},
|
| 1036 |
+
{
|
| 1037 |
+
"epoch": 4.8169556840077075,
|
| 1038 |
+
"grad_norm": 0.5916352272033691,
|
| 1039 |
+
"learning_rate": 3.95546385122461e-05,
|
| 1040 |
+
"loss": 0.1325,
|
| 1041 |
+
"step": 60000
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 4.857096981374438,
|
| 1045 |
+
"grad_norm": 3.2611584663391113,
|
| 1046 |
+
"learning_rate": 3.8214802508173e-05,
|
| 1047 |
+
"loss": 0.126,
|
| 1048 |
+
"step": 60500
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 4.897238278741169,
|
| 1052 |
+
"grad_norm": 2.499826192855835,
|
| 1053 |
+
"learning_rate": 3.68749665040999e-05,
|
| 1054 |
+
"loss": 0.127,
|
| 1055 |
+
"step": 61000
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 4.9373795761079,
|
| 1059 |
+
"grad_norm": 3.263596773147583,
|
| 1060 |
+
"learning_rate": 3.55351305000268e-05,
|
| 1061 |
+
"loss": 0.1263,
|
| 1062 |
+
"step": 61500
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 4.977520873474631,
|
| 1066 |
+
"grad_norm": 3.2504055500030518,
|
| 1067 |
+
"learning_rate": 3.419529449595369e-05,
|
| 1068 |
+
"loss": 0.1298,
|
| 1069 |
+
"step": 62000
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 5.001605651894669,
|
| 1073 |
+
"eval_loss": 0.13082526624202728,
|
| 1074 |
+
"eval_runtime": 9.18,
|
| 1075 |
+
"eval_samples_per_second": 10.893,
|
| 1076 |
+
"eval_steps_per_second": 5.447,
|
| 1077 |
+
"step": 62300
|
| 1078 |
+
},
|
| 1079 |
+
{
|
| 1080 |
+
"epoch": 5.017662170841362,
|
| 1081 |
+
"grad_norm": 1.3521040678024292,
|
| 1082 |
+
"learning_rate": 3.28554584918806e-05,
|
| 1083 |
+
"loss": 0.0973,
|
| 1084 |
+
"step": 62500
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 5.057803468208093,
|
| 1088 |
+
"grad_norm": 1.273701786994934,
|
| 1089 |
+
"learning_rate": 3.151830215981564e-05,
|
| 1090 |
+
"loss": 0.1059,
|
| 1091 |
+
"step": 63000
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 5.097944765574823,
|
| 1095 |
+
"grad_norm": 3.1900885105133057,
|
| 1096 |
+
"learning_rate": 3.017846615574254e-05,
|
| 1097 |
+
"loss": 0.1095,
|
| 1098 |
+
"step": 63500
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 5.138086062941555,
|
| 1102 |
+
"grad_norm": 1.1862250566482544,
|
| 1103 |
+
"learning_rate": 2.883863015166944e-05,
|
| 1104 |
+
"loss": 0.1053,
|
| 1105 |
+
"step": 64000
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 5.178227360308285,
|
| 1109 |
+
"grad_norm": 0.06329891085624695,
|
| 1110 |
+
"learning_rate": 2.7498794147596337e-05,
|
| 1111 |
+
"loss": 0.1064,
|
| 1112 |
+
"step": 64500
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 5.201669877970456,
|
| 1116 |
+
"eval_loss": 0.12689721584320068,
|
| 1117 |
+
"eval_runtime": 9.2645,
|
| 1118 |
+
"eval_samples_per_second": 10.794,
|
| 1119 |
+
"eval_steps_per_second": 5.397,
|
| 1120 |
+
"step": 64792
|
| 1121 |
+
},
|
| 1122 |
+
{
|
| 1123 |
+
"epoch": 5.218368657675016,
|
| 1124 |
+
"grad_norm": 1.3881759643554688,
|
| 1125 |
+
"learning_rate": 2.615895814352323e-05,
|
| 1126 |
+
"loss": 0.0983,
|
| 1127 |
+
"step": 65000
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 5.258509955041747,
|
| 1131 |
+
"grad_norm": 2.040412664413452,
|
| 1132 |
+
"learning_rate": 2.4819122139450132e-05,
|
| 1133 |
+
"loss": 0.11,
|
| 1134 |
+
"step": 65500
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 5.298651252408478,
|
| 1138 |
+
"grad_norm": 1.5292569398880005,
|
| 1139 |
+
"learning_rate": 2.348196580738518e-05,
|
| 1140 |
+
"loss": 0.1087,
|
| 1141 |
+
"step": 66000
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 5.338792549775209,
|
| 1145 |
+
"grad_norm": 1.2713596820831299,
|
| 1146 |
+
"learning_rate": 2.2142129803312077e-05,
|
| 1147 |
+
"loss": 0.1082,
|
| 1148 |
+
"step": 66500
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 5.37893384714194,
|
| 1152 |
+
"grad_norm": 1.5126820802688599,
|
| 1153 |
+
"learning_rate": 2.080497347124712e-05,
|
| 1154 |
+
"loss": 0.1063,
|
| 1155 |
+
"step": 67000
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 5.401734104046243,
|
| 1159 |
+
"eval_loss": 0.12622660398483276,
|
| 1160 |
+
"eval_runtime": 9.2201,
|
| 1161 |
+
"eval_samples_per_second": 10.846,
|
| 1162 |
+
"eval_steps_per_second": 5.423,
|
| 1163 |
+
"step": 67284
|
| 1164 |
+
},
|
| 1165 |
+
{
|
| 1166 |
+
"epoch": 5.41907514450867,
|
| 1167 |
+
"grad_norm": 1.634347915649414,
|
| 1168 |
+
"learning_rate": 1.9465137467174018e-05,
|
| 1169 |
+
"loss": 0.1076,
|
| 1170 |
+
"step": 67500
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 5.459216441875402,
|
| 1174 |
+
"grad_norm": 1.4614862203598022,
|
| 1175 |
+
"learning_rate": 1.8125301463100915e-05,
|
| 1176 |
+
"loss": 0.1001,
|
| 1177 |
+
"step": 68000
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 5.499357739242132,
|
| 1181 |
+
"grad_norm": 1.3177045583724976,
|
| 1182 |
+
"learning_rate": 1.6785465459027816e-05,
|
| 1183 |
+
"loss": 0.1041,
|
| 1184 |
+
"step": 68500
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 5.539499036608863,
|
| 1188 |
+
"grad_norm": 2.3282182216644287,
|
| 1189 |
+
"learning_rate": 1.5445629454954714e-05,
|
| 1190 |
+
"loss": 0.1021,
|
| 1191 |
+
"step": 69000
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 5.579640333975594,
|
| 1195 |
+
"grad_norm": 1.7429168224334717,
|
| 1196 |
+
"learning_rate": 1.4105793450881613e-05,
|
| 1197 |
+
"loss": 0.1036,
|
| 1198 |
+
"step": 69500
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 5.60179833012203,
|
| 1202 |
+
"eval_loss": 0.12125935405492783,
|
| 1203 |
+
"eval_runtime": 9.2288,
|
| 1204 |
+
"eval_samples_per_second": 10.836,
|
| 1205 |
+
"eval_steps_per_second": 5.418,
|
| 1206 |
+
"step": 69776
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 5.619781631342325,
|
| 1210 |
+
"grad_norm": 0.19996996223926544,
|
| 1211 |
+
"learning_rate": 1.2765957446808511e-05,
|
| 1212 |
+
"loss": 0.098,
|
| 1213 |
+
"step": 70000
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 5.659922928709056,
|
| 1217 |
+
"grad_norm": 3.154064893722534,
|
| 1218 |
+
"learning_rate": 1.142612144273541e-05,
|
| 1219 |
+
"loss": 0.101,
|
| 1220 |
+
"step": 70500
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 5.700064226075787,
|
| 1224 |
+
"grad_norm": 0.9969759583473206,
|
| 1225 |
+
"learning_rate": 1.0086285438662308e-05,
|
| 1226 |
+
"loss": 0.1043,
|
| 1227 |
+
"step": 71000
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 5.740205523442517,
|
| 1231 |
+
"grad_norm": 0.29482555389404297,
|
| 1232 |
+
"learning_rate": 8.749129106597353e-06,
|
| 1233 |
+
"loss": 0.0978,
|
| 1234 |
+
"step": 71500
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 5.780346820809249,
|
| 1238 |
+
"grad_norm": 7.377176761627197,
|
| 1239 |
+
"learning_rate": 7.409293102524252e-06,
|
| 1240 |
+
"loss": 0.1084,
|
| 1241 |
+
"step": 72000
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 5.8018625561978165,
|
| 1245 |
+
"eval_loss": 0.11656199395656586,
|
| 1246 |
+
"eval_runtime": 9.2458,
|
| 1247 |
+
"eval_samples_per_second": 10.816,
|
| 1248 |
+
"eval_steps_per_second": 5.408,
|
| 1249 |
+
"step": 72268
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 5.820488118175979,
|
| 1253 |
+
"grad_norm": 1.861075520515442,
|
| 1254 |
+
"learning_rate": 6.06945709845115e-06,
|
| 1255 |
+
"loss": 0.1045,
|
| 1256 |
+
"step": 72500
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 5.860629415542711,
|
| 1260 |
+
"grad_norm": 3.3500561714172363,
|
| 1261 |
+
"learning_rate": 4.729621094378049e-06,
|
| 1262 |
+
"loss": 0.103,
|
| 1263 |
+
"step": 73000
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 5.900770712909441,
|
| 1267 |
+
"grad_norm": 0.29247429966926575,
|
| 1268 |
+
"learning_rate": 3.3924647623130934e-06,
|
| 1269 |
+
"loss": 0.1022,
|
| 1270 |
+
"step": 73500
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 5.940912010276172,
|
| 1274 |
+
"grad_norm": 2.5670969486236572,
|
| 1275 |
+
"learning_rate": 2.0526287582399915e-06,
|
| 1276 |
+
"loss": 0.1048,
|
| 1277 |
+
"step": 74000
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 5.981053307642903,
|
| 1281 |
+
"grad_norm": 1.4825931787490845,
|
| 1282 |
+
"learning_rate": 7.1279275416689e-07,
|
| 1283 |
+
"loss": 0.1009,
|
| 1284 |
+
"step": 74500
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 6.0,
|
| 1288 |
+
"step": 74736,
|
| 1289 |
+
"total_flos": 1.1033260642249114e+18,
|
| 1290 |
+
"train_loss": 0.01737067021080218,
|
| 1291 |
+
"train_runtime": 11655.5312,
|
| 1292 |
+
"train_samples_per_second": 12.824,
|
| 1293 |
+
"train_steps_per_second": 6.412
|
| 1294 |
+
}
|
| 1295 |
+
],
|
| 1296 |
+
"logging_steps": 500,
|
| 1297 |
+
"max_steps": 74736,
|
| 1298 |
+
"num_input_tokens_seen": 0,
|
| 1299 |
+
"num_train_epochs": 6,
|
| 1300 |
+
"save_steps": 500,
|
| 1301 |
+
"stateful_callbacks": {
|
| 1302 |
+
"TrainerControl": {
|
| 1303 |
+
"args": {
|
| 1304 |
+
"should_epoch_stop": false,
|
| 1305 |
+
"should_evaluate": false,
|
| 1306 |
+
"should_log": false,
|
| 1307 |
+
"should_save": true,
|
| 1308 |
+
"should_training_stop": true
|
| 1309 |
+
},
|
| 1310 |
+
"attributes": {}
|
| 1311 |
+
}
|
| 1312 |
+
},
|
| 1313 |
+
"total_flos": 1.1033260642249114e+18,
|
| 1314 |
+
"train_batch_size": 2,
|
| 1315 |
+
"trial_name": null,
|
| 1316 |
+
"trial_params": null
|
| 1317 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b479fe1ba21f33b0b5e546f065f753fb4c3b393f01a78db43f67de7f31aa319d
|
| 3 |
+
size 5432
|