End of training
Browse files- README.md +88 -0
- adapter_config.json +32 -0
- adapter_model.safetensors +3 -0
- trainer_state.json +1324 -0
- training_args.bin +3 -0
README.md
ADDED
|
@@ -0,0 +1,88 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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-task1-more_symbols_template_small-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-task1-more_symbols_template_small-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: 0.1504
|
| 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: 4
|
| 40 |
+
- eval_batch_size: 4
|
| 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 |
+
- num_epochs: 6
|
| 45 |
+
- mixed_precision_training: Native AMP
|
| 46 |
+
|
| 47 |
+
### Training results
|
| 48 |
+
|
| 49 |
+
| Training Loss | Epoch | Step | Validation Loss |
|
| 50 |
+
|:-------------:|:------:|:-----:|:---------------:|
|
| 51 |
+
| 0.3464 | 0.2000 | 2515 | 0.3264 |
|
| 52 |
+
| 0.2953 | 0.4001 | 5030 | 0.2775 |
|
| 53 |
+
| 0.2629 | 0.6001 | 7545 | 0.2489 |
|
| 54 |
+
| 0.2453 | 0.8001 | 10060 | 0.2290 |
|
| 55 |
+
| 0.2283 | 1.0002 | 12575 | 0.2216 |
|
| 56 |
+
| 0.2102 | 1.2002 | 15090 | 0.2148 |
|
| 57 |
+
| 0.2057 | 1.4002 | 17605 | 0.2071 |
|
| 58 |
+
| 0.1965 | 1.6003 | 20120 | 0.1968 |
|
| 59 |
+
| 0.185 | 1.8003 | 22635 | 0.1940 |
|
| 60 |
+
| 0.1894 | 2.0003 | 25150 | 0.1880 |
|
| 61 |
+
| 0.159 | 2.2003 | 27665 | 0.1870 |
|
| 62 |
+
| 0.1583 | 2.4004 | 30180 | 0.1786 |
|
| 63 |
+
| 0.1592 | 2.6004 | 32695 | 0.1774 |
|
| 64 |
+
| 0.1609 | 2.8004 | 35210 | 0.1720 |
|
| 65 |
+
| 0.1571 | 3.0005 | 37725 | 0.1670 |
|
| 66 |
+
| 0.1301 | 3.2005 | 40240 | 0.1694 |
|
| 67 |
+
| 0.1362 | 3.4005 | 42755 | 0.1690 |
|
| 68 |
+
| 0.1384 | 3.6006 | 45270 | 0.1595 |
|
| 69 |
+
| 0.1296 | 3.8006 | 47785 | 0.1582 |
|
| 70 |
+
| 0.1332 | 4.0006 | 50300 | 0.1557 |
|
| 71 |
+
| 0.1122 | 4.2007 | 52815 | 0.1581 |
|
| 72 |
+
| 0.1116 | 4.4007 | 55330 | 0.1538 |
|
| 73 |
+
| 0.1093 | 4.6007 | 57845 | 0.1525 |
|
| 74 |
+
| 0.1082 | 4.8008 | 60360 | 0.1534 |
|
| 75 |
+
| 0.1056 | 5.0008 | 62875 | 0.1500 |
|
| 76 |
+
| 0.0904 | 5.2008 | 65390 | 0.1529 |
|
| 77 |
+
| 0.0901 | 5.4009 | 67905 | 0.1516 |
|
| 78 |
+
| 0.0893 | 5.6009 | 70420 | 0.1489 |
|
| 79 |
+
| 0.0928 | 5.8009 | 72935 | 0.1504 |
|
| 80 |
+
|
| 81 |
+
|
| 82 |
+
### Framework versions
|
| 83 |
+
|
| 84 |
+
- PEFT 0.14.0
|
| 85 |
+
- Transformers 4.47.0
|
| 86 |
+
- Pytorch 2.5.1+cu124
|
| 87 |
+
- Datasets 3.2.0
|
| 88 |
+
- 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 sha256:56f11dac150eeb3ed6ce763b1331670f9d5158820457b2afe6785284e6458160
|
| 3 |
+
size 6304096
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1324 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 6.0,
|
| 5 |
+
"eval_steps": 2515,
|
| 6 |
+
"global_step": 75438,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.03976775630318937,
|
| 13 |
+
"grad_norm": 1.5738416910171509,
|
| 14 |
+
"learning_rate": 0.00019867440812322703,
|
| 15 |
+
"loss": 0.5177,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.07953551260637874,
|
| 20 |
+
"grad_norm": 1.153595209121704,
|
| 21 |
+
"learning_rate": 0.00019734881624645405,
|
| 22 |
+
"loss": 0.4158,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.11930326890956812,
|
| 27 |
+
"grad_norm": 1.121400237083435,
|
| 28 |
+
"learning_rate": 0.00019602322436968107,
|
| 29 |
+
"loss": 0.3828,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.15907102521275748,
|
| 34 |
+
"grad_norm": 1.9384912252426147,
|
| 35 |
+
"learning_rate": 0.0001946976324929081,
|
| 36 |
+
"loss": 0.3472,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.19883878151594686,
|
| 41 |
+
"grad_norm": 1.5707042217254639,
|
| 42 |
+
"learning_rate": 0.00019337469179988865,
|
| 43 |
+
"loss": 0.3464,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.20003181420504254,
|
| 48 |
+
"eval_loss": 0.3264305293560028,
|
| 49 |
+
"eval_runtime": 45.4813,
|
| 50 |
+
"eval_samples_per_second": 10.994,
|
| 51 |
+
"eval_steps_per_second": 2.748,
|
| 52 |
+
"step": 2515
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.23860653781913624,
|
| 56 |
+
"grad_norm": 1.2760212421417236,
|
| 57 |
+
"learning_rate": 0.00019204909992311567,
|
| 58 |
+
"loss": 0.3264,
|
| 59 |
+
"step": 3000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.2783742941223256,
|
| 63 |
+
"grad_norm": 1.840829849243164,
|
| 64 |
+
"learning_rate": 0.0001907235080463427,
|
| 65 |
+
"loss": 0.3098,
|
| 66 |
+
"step": 3500
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.31814205042551497,
|
| 70 |
+
"grad_norm": 0.8722560405731201,
|
| 71 |
+
"learning_rate": 0.0001893979161695697,
|
| 72 |
+
"loss": 0.3032,
|
| 73 |
+
"step": 4000
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.35790980672870437,
|
| 77 |
+
"grad_norm": 1.4368563890457153,
|
| 78 |
+
"learning_rate": 0.00018807232429279676,
|
| 79 |
+
"loss": 0.2941,
|
| 80 |
+
"step": 4500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.3976775630318937,
|
| 84 |
+
"grad_norm": 4.807660102844238,
|
| 85 |
+
"learning_rate": 0.00018675203478353084,
|
| 86 |
+
"loss": 0.2953,
|
| 87 |
+
"step": 5000
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.4000636284100851,
|
| 91 |
+
"eval_loss": 0.2775295376777649,
|
| 92 |
+
"eval_runtime": 45.1658,
|
| 93 |
+
"eval_samples_per_second": 11.07,
|
| 94 |
+
"eval_steps_per_second": 2.768,
|
| 95 |
+
"step": 5030
|
| 96 |
+
},
|
| 97 |
+
{
|
| 98 |
+
"epoch": 0.4374453193350831,
|
| 99 |
+
"grad_norm": 2.1536879539489746,
|
| 100 |
+
"learning_rate": 0.00018542644290675788,
|
| 101 |
+
"loss": 0.2842,
|
| 102 |
+
"step": 5500
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 0.4772130756382725,
|
| 106 |
+
"grad_norm": 1.3601027727127075,
|
| 107 |
+
"learning_rate": 0.00018410085102998488,
|
| 108 |
+
"loss": 0.2827,
|
| 109 |
+
"step": 6000
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 0.5169808319414618,
|
| 113 |
+
"grad_norm": 1.182559847831726,
|
| 114 |
+
"learning_rate": 0.00018277525915321192,
|
| 115 |
+
"loss": 0.2763,
|
| 116 |
+
"step": 6500
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.5567485882446512,
|
| 120 |
+
"grad_norm": 0.9382211565971375,
|
| 121 |
+
"learning_rate": 0.00018144966727643894,
|
| 122 |
+
"loss": 0.2662,
|
| 123 |
+
"step": 7000
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 0.5965163445478406,
|
| 127 |
+
"grad_norm": 1.8462032079696655,
|
| 128 |
+
"learning_rate": 0.0001801267265834195,
|
| 129 |
+
"loss": 0.2629,
|
| 130 |
+
"step": 7500
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.6000954426151277,
|
| 134 |
+
"eval_loss": 0.24887844920158386,
|
| 135 |
+
"eval_runtime": 45.2464,
|
| 136 |
+
"eval_samples_per_second": 11.051,
|
| 137 |
+
"eval_steps_per_second": 2.763,
|
| 138 |
+
"step": 7545
|
| 139 |
+
},
|
| 140 |
+
{
|
| 141 |
+
"epoch": 0.6362841008510299,
|
| 142 |
+
"grad_norm": 1.0685694217681885,
|
| 143 |
+
"learning_rate": 0.00017880113470664652,
|
| 144 |
+
"loss": 0.2539,
|
| 145 |
+
"step": 8000
|
| 146 |
+
},
|
| 147 |
+
{
|
| 148 |
+
"epoch": 0.6760518571542193,
|
| 149 |
+
"grad_norm": 1.1819478273391724,
|
| 150 |
+
"learning_rate": 0.00017747554282987354,
|
| 151 |
+
"loss": 0.2587,
|
| 152 |
+
"step": 8500
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.7158196134574087,
|
| 156 |
+
"grad_norm": 1.314528465270996,
|
| 157 |
+
"learning_rate": 0.00017614995095310056,
|
| 158 |
+
"loss": 0.2539,
|
| 159 |
+
"step": 9000
|
| 160 |
+
},
|
| 161 |
+
{
|
| 162 |
+
"epoch": 0.7555873697605981,
|
| 163 |
+
"grad_norm": 1.265201210975647,
|
| 164 |
+
"learning_rate": 0.00017482435907632758,
|
| 165 |
+
"loss": 0.2468,
|
| 166 |
+
"step": 9500
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 0.7953551260637874,
|
| 170 |
+
"grad_norm": 1.079946756362915,
|
| 171 |
+
"learning_rate": 0.0001734987671995546,
|
| 172 |
+
"loss": 0.2453,
|
| 173 |
+
"step": 10000
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.8001272568201702,
|
| 177 |
+
"eval_loss": 0.22904042899608612,
|
| 178 |
+
"eval_runtime": 45.6945,
|
| 179 |
+
"eval_samples_per_second": 10.942,
|
| 180 |
+
"eval_steps_per_second": 2.736,
|
| 181 |
+
"step": 10060
|
| 182 |
+
},
|
| 183 |
+
{
|
| 184 |
+
"epoch": 0.8351228823669768,
|
| 185 |
+
"grad_norm": 0.8490906357765198,
|
| 186 |
+
"learning_rate": 0.00017217317532278162,
|
| 187 |
+
"loss": 0.2428,
|
| 188 |
+
"step": 10500
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.8748906386701663,
|
| 192 |
+
"grad_norm": 0.9842996597290039,
|
| 193 |
+
"learning_rate": 0.00017084758344600864,
|
| 194 |
+
"loss": 0.2316,
|
| 195 |
+
"step": 11000
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 0.9146583949733555,
|
| 199 |
+
"grad_norm": 1.168321967124939,
|
| 200 |
+
"learning_rate": 0.00016952464275298923,
|
| 201 |
+
"loss": 0.2371,
|
| 202 |
+
"step": 11500
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 0.954426151276545,
|
| 206 |
+
"grad_norm": 1.088979959487915,
|
| 207 |
+
"learning_rate": 0.00016819905087621622,
|
| 208 |
+
"loss": 0.2428,
|
| 209 |
+
"step": 12000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 0.9941939075797344,
|
| 213 |
+
"grad_norm": 1.175786018371582,
|
| 214 |
+
"learning_rate": 0.00016687345899944327,
|
| 215 |
+
"loss": 0.2283,
|
| 216 |
+
"step": 12500
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 1.0001590710252128,
|
| 220 |
+
"eval_loss": 0.2215883731842041,
|
| 221 |
+
"eval_runtime": 45.2828,
|
| 222 |
+
"eval_samples_per_second": 11.042,
|
| 223 |
+
"eval_steps_per_second": 2.76,
|
| 224 |
+
"step": 12575
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 1.0339616638829237,
|
| 228 |
+
"grad_norm": 1.0268359184265137,
|
| 229 |
+
"learning_rate": 0.00016554786712267026,
|
| 230 |
+
"loss": 0.209,
|
| 231 |
+
"step": 13000
|
| 232 |
+
},
|
| 233 |
+
{
|
| 234 |
+
"epoch": 1.0737294201861132,
|
| 235 |
+
"grad_norm": 1.1008212566375732,
|
| 236 |
+
"learning_rate": 0.0001642222752458973,
|
| 237 |
+
"loss": 0.212,
|
| 238 |
+
"step": 13500
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 1.1134971764893025,
|
| 242 |
+
"grad_norm": 1.6737862825393677,
|
| 243 |
+
"learning_rate": 0.0001628966833691243,
|
| 244 |
+
"loss": 0.2189,
|
| 245 |
+
"step": 14000
|
| 246 |
+
},
|
| 247 |
+
{
|
| 248 |
+
"epoch": 1.1532649327924918,
|
| 249 |
+
"grad_norm": 1.5881727933883667,
|
| 250 |
+
"learning_rate": 0.00016157374267610487,
|
| 251 |
+
"loss": 0.2136,
|
| 252 |
+
"step": 14500
|
| 253 |
+
},
|
| 254 |
+
{
|
| 255 |
+
"epoch": 1.1930326890956813,
|
| 256 |
+
"grad_norm": 2.8963282108306885,
|
| 257 |
+
"learning_rate": 0.0001602481507993319,
|
| 258 |
+
"loss": 0.2102,
|
| 259 |
+
"step": 15000
|
| 260 |
+
},
|
| 261 |
+
{
|
| 262 |
+
"epoch": 1.2001908852302554,
|
| 263 |
+
"eval_loss": 0.2148231863975525,
|
| 264 |
+
"eval_runtime": 45.2446,
|
| 265 |
+
"eval_samples_per_second": 11.051,
|
| 266 |
+
"eval_steps_per_second": 2.763,
|
| 267 |
+
"step": 15090
|
| 268 |
+
},
|
| 269 |
+
{
|
| 270 |
+
"epoch": 1.2328004453988706,
|
| 271 |
+
"grad_norm": 2.0356953144073486,
|
| 272 |
+
"learning_rate": 0.00015892255892255893,
|
| 273 |
+
"loss": 0.2057,
|
| 274 |
+
"step": 15500
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 1.27256820170206,
|
| 278 |
+
"grad_norm": 1.0987403392791748,
|
| 279 |
+
"learning_rate": 0.00015759696704578595,
|
| 280 |
+
"loss": 0.2122,
|
| 281 |
+
"step": 16000
|
| 282 |
+
},
|
| 283 |
+
{
|
| 284 |
+
"epoch": 1.3123359580052494,
|
| 285 |
+
"grad_norm": 1.197990894317627,
|
| 286 |
+
"learning_rate": 0.00015627137516901297,
|
| 287 |
+
"loss": 0.2055,
|
| 288 |
+
"step": 16500
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 1.3521037143084387,
|
| 292 |
+
"grad_norm": 1.2933690547943115,
|
| 293 |
+
"learning_rate": 0.00015494578329224,
|
| 294 |
+
"loss": 0.2056,
|
| 295 |
+
"step": 17000
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 1.391871470611628,
|
| 299 |
+
"grad_norm": 1.2850003242492676,
|
| 300 |
+
"learning_rate": 0.00015362284259922055,
|
| 301 |
+
"loss": 0.2057,
|
| 302 |
+
"step": 17500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 1.400222699435298,
|
| 306 |
+
"eval_loss": 0.20713141560554504,
|
| 307 |
+
"eval_runtime": 45.4485,
|
| 308 |
+
"eval_samples_per_second": 11.001,
|
| 309 |
+
"eval_steps_per_second": 2.75,
|
| 310 |
+
"step": 17605
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 1.4316392269148175,
|
| 314 |
+
"grad_norm": 0.9641985297203064,
|
| 315 |
+
"learning_rate": 0.00015229725072244757,
|
| 316 |
+
"loss": 0.2078,
|
| 317 |
+
"step": 18000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 1.4714069832180068,
|
| 321 |
+
"grad_norm": 1.390886902809143,
|
| 322 |
+
"learning_rate": 0.0001509716588456746,
|
| 323 |
+
"loss": 0.1979,
|
| 324 |
+
"step": 18500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 1.5111747395211963,
|
| 328 |
+
"grad_norm": 2.135011672973633,
|
| 329 |
+
"learning_rate": 0.0001496460669689016,
|
| 330 |
+
"loss": 0.2001,
|
| 331 |
+
"step": 19000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 1.5509424958243856,
|
| 335 |
+
"grad_norm": 3.0572569370269775,
|
| 336 |
+
"learning_rate": 0.00014832312627588218,
|
| 337 |
+
"loss": 0.1954,
|
| 338 |
+
"step": 19500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.5907102521275749,
|
| 342 |
+
"grad_norm": 1.4685252904891968,
|
| 343 |
+
"learning_rate": 0.00014699753439910922,
|
| 344 |
+
"loss": 0.1965,
|
| 345 |
+
"step": 20000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 1.6002545136403405,
|
| 349 |
+
"eval_loss": 0.19675177335739136,
|
| 350 |
+
"eval_runtime": 45.6648,
|
| 351 |
+
"eval_samples_per_second": 10.949,
|
| 352 |
+
"eval_steps_per_second": 2.737,
|
| 353 |
+
"step": 20120
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.6304780084307642,
|
| 357 |
+
"grad_norm": 1.2274647951126099,
|
| 358 |
+
"learning_rate": 0.00014567194252233621,
|
| 359 |
+
"loss": 0.1924,
|
| 360 |
+
"step": 20500
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 1.6702457647339537,
|
| 364 |
+
"grad_norm": 1.2652219533920288,
|
| 365 |
+
"learning_rate": 0.00014434635064556326,
|
| 366 |
+
"loss": 0.1962,
|
| 367 |
+
"step": 21000
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 1.7100135210371432,
|
| 371 |
+
"grad_norm": 1.625205636024475,
|
| 372 |
+
"learning_rate": 0.00014302075876879025,
|
| 373 |
+
"loss": 0.1965,
|
| 374 |
+
"step": 21500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.7497812773403325,
|
| 378 |
+
"grad_norm": 2.484333038330078,
|
| 379 |
+
"learning_rate": 0.0001416951668920173,
|
| 380 |
+
"loss": 0.1982,
|
| 381 |
+
"step": 22000
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 1.7895490336435218,
|
| 385 |
+
"grad_norm": 0.98336261510849,
|
| 386 |
+
"learning_rate": 0.00014037222619899786,
|
| 387 |
+
"loss": 0.185,
|
| 388 |
+
"step": 22500
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 1.8002863278453831,
|
| 392 |
+
"eval_loss": 0.1940314620733261,
|
| 393 |
+
"eval_runtime": 45.1786,
|
| 394 |
+
"eval_samples_per_second": 11.067,
|
| 395 |
+
"eval_steps_per_second": 2.767,
|
| 396 |
+
"step": 22635
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 1.829316789946711,
|
| 400 |
+
"grad_norm": 1.6227530241012573,
|
| 401 |
+
"learning_rate": 0.00013904663432222488,
|
| 402 |
+
"loss": 0.1872,
|
| 403 |
+
"step": 23000
|
| 404 |
+
},
|
| 405 |
+
{
|
| 406 |
+
"epoch": 1.8690845462499006,
|
| 407 |
+
"grad_norm": 1.9819400310516357,
|
| 408 |
+
"learning_rate": 0.0001377210424454519,
|
| 409 |
+
"loss": 0.1941,
|
| 410 |
+
"step": 23500
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 1.90885230255309,
|
| 414 |
+
"grad_norm": 0.8801836967468262,
|
| 415 |
+
"learning_rate": 0.00013639545056867892,
|
| 416 |
+
"loss": 0.1849,
|
| 417 |
+
"step": 24000
|
| 418 |
+
},
|
| 419 |
+
{
|
| 420 |
+
"epoch": 1.9486200588562794,
|
| 421 |
+
"grad_norm": 1.538822889328003,
|
| 422 |
+
"learning_rate": 0.00013506985869190594,
|
| 423 |
+
"loss": 0.1915,
|
| 424 |
+
"step": 24500
|
| 425 |
+
},
|
| 426 |
+
{
|
| 427 |
+
"epoch": 1.9883878151594687,
|
| 428 |
+
"grad_norm": 1.2602678537368774,
|
| 429 |
+
"learning_rate": 0.0001337469179988865,
|
| 430 |
+
"loss": 0.1894,
|
| 431 |
+
"step": 25000
|
| 432 |
+
},
|
| 433 |
+
{
|
| 434 |
+
"epoch": 2.0003181420504257,
|
| 435 |
+
"eval_loss": 0.18800613284111023,
|
| 436 |
+
"eval_runtime": 45.2739,
|
| 437 |
+
"eval_samples_per_second": 11.044,
|
| 438 |
+
"eval_steps_per_second": 2.761,
|
| 439 |
+
"step": 25150
|
| 440 |
+
},
|
| 441 |
+
{
|
| 442 |
+
"epoch": 2.028155571462658,
|
| 443 |
+
"grad_norm": 1.2077054977416992,
|
| 444 |
+
"learning_rate": 0.00013242132612211352,
|
| 445 |
+
"loss": 0.172,
|
| 446 |
+
"step": 25500
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 2.0679233277658473,
|
| 450 |
+
"grad_norm": 1.3468750715255737,
|
| 451 |
+
"learning_rate": 0.00013109573424534054,
|
| 452 |
+
"loss": 0.168,
|
| 453 |
+
"step": 26000
|
| 454 |
+
},
|
| 455 |
+
{
|
| 456 |
+
"epoch": 2.107691084069037,
|
| 457 |
+
"grad_norm": 1.2310609817504883,
|
| 458 |
+
"learning_rate": 0.00012977014236856756,
|
| 459 |
+
"loss": 0.165,
|
| 460 |
+
"step": 26500
|
| 461 |
+
},
|
| 462 |
+
{
|
| 463 |
+
"epoch": 2.1474588403722263,
|
| 464 |
+
"grad_norm": 1.662156105041504,
|
| 465 |
+
"learning_rate": 0.00012844720167554813,
|
| 466 |
+
"loss": 0.1694,
|
| 467 |
+
"step": 27000
|
| 468 |
+
},
|
| 469 |
+
{
|
| 470 |
+
"epoch": 2.1872265966754156,
|
| 471 |
+
"grad_norm": 1.1639182567596436,
|
| 472 |
+
"learning_rate": 0.00012712160979877517,
|
| 473 |
+
"loss": 0.159,
|
| 474 |
+
"step": 27500
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 2.2003499562554683,
|
| 478 |
+
"eval_loss": 0.1869928389787674,
|
| 479 |
+
"eval_runtime": 45.1579,
|
| 480 |
+
"eval_samples_per_second": 11.072,
|
| 481 |
+
"eval_steps_per_second": 2.768,
|
| 482 |
+
"step": 27665
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 2.226994352978605,
|
| 486 |
+
"grad_norm": 1.9803431034088135,
|
| 487 |
+
"learning_rate": 0.00012579601792200217,
|
| 488 |
+
"loss": 0.1675,
|
| 489 |
+
"step": 28000
|
| 490 |
+
},
|
| 491 |
+
{
|
| 492 |
+
"epoch": 2.2667621092817942,
|
| 493 |
+
"grad_norm": 2.7467844486236572,
|
| 494 |
+
"learning_rate": 0.0001244704260452292,
|
| 495 |
+
"loss": 0.1701,
|
| 496 |
+
"step": 28500
|
| 497 |
+
},
|
| 498 |
+
{
|
| 499 |
+
"epoch": 2.3065298655849835,
|
| 500 |
+
"grad_norm": 1.4122605323791504,
|
| 501 |
+
"learning_rate": 0.00012314748535220978,
|
| 502 |
+
"loss": 0.17,
|
| 503 |
+
"step": 29000
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 2.3462976218881733,
|
| 507 |
+
"grad_norm": 1.572930097579956,
|
| 508 |
+
"learning_rate": 0.0001218218934754368,
|
| 509 |
+
"loss": 0.1752,
|
| 510 |
+
"step": 29500
|
| 511 |
+
},
|
| 512 |
+
{
|
| 513 |
+
"epoch": 2.3860653781913626,
|
| 514 |
+
"grad_norm": 0.8454307317733765,
|
| 515 |
+
"learning_rate": 0.0001204963015986638,
|
| 516 |
+
"loss": 0.1583,
|
| 517 |
+
"step": 30000
|
| 518 |
+
},
|
| 519 |
+
{
|
| 520 |
+
"epoch": 2.400381770460511,
|
| 521 |
+
"eval_loss": 0.17864643037319183,
|
| 522 |
+
"eval_runtime": 45.3939,
|
| 523 |
+
"eval_samples_per_second": 11.015,
|
| 524 |
+
"eval_steps_per_second": 2.754,
|
| 525 |
+
"step": 30180
|
| 526 |
+
},
|
| 527 |
+
{
|
| 528 |
+
"epoch": 2.425833134494552,
|
| 529 |
+
"grad_norm": 1.514496088027954,
|
| 530 |
+
"learning_rate": 0.00011917070972189083,
|
| 531 |
+
"loss": 0.1625,
|
| 532 |
+
"step": 30500
|
| 533 |
+
},
|
| 534 |
+
{
|
| 535 |
+
"epoch": 2.465600890797741,
|
| 536 |
+
"grad_norm": 1.4132134914398193,
|
| 537 |
+
"learning_rate": 0.00011784511784511784,
|
| 538 |
+
"loss": 0.1646,
|
| 539 |
+
"step": 31000
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 2.5053686471009304,
|
| 543 |
+
"grad_norm": 1.2174311876296997,
|
| 544 |
+
"learning_rate": 0.00011651952596834487,
|
| 545 |
+
"loss": 0.1641,
|
| 546 |
+
"step": 31500
|
| 547 |
+
},
|
| 548 |
+
{
|
| 549 |
+
"epoch": 2.54513640340412,
|
| 550 |
+
"grad_norm": 1.184470772743225,
|
| 551 |
+
"learning_rate": 0.00011519393409157189,
|
| 552 |
+
"loss": 0.1614,
|
| 553 |
+
"step": 32000
|
| 554 |
+
},
|
| 555 |
+
{
|
| 556 |
+
"epoch": 2.5849041597073095,
|
| 557 |
+
"grad_norm": 2.204496145248413,
|
| 558 |
+
"learning_rate": 0.00011387099339855246,
|
| 559 |
+
"loss": 0.1592,
|
| 560 |
+
"step": 32500
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 2.600413584665553,
|
| 564 |
+
"eval_loss": 0.17739520967006683,
|
| 565 |
+
"eval_runtime": 45.4474,
|
| 566 |
+
"eval_samples_per_second": 11.002,
|
| 567 |
+
"eval_steps_per_second": 2.75,
|
| 568 |
+
"step": 32695
|
| 569 |
+
},
|
| 570 |
+
{
|
| 571 |
+
"epoch": 2.6246719160104988,
|
| 572 |
+
"grad_norm": 1.5031511783599854,
|
| 573 |
+
"learning_rate": 0.00011254540152177948,
|
| 574 |
+
"loss": 0.1654,
|
| 575 |
+
"step": 33000
|
| 576 |
+
},
|
| 577 |
+
{
|
| 578 |
+
"epoch": 2.664439672313688,
|
| 579 |
+
"grad_norm": 2.9933559894561768,
|
| 580 |
+
"learning_rate": 0.0001112198096450065,
|
| 581 |
+
"loss": 0.1604,
|
| 582 |
+
"step": 33500
|
| 583 |
+
},
|
| 584 |
+
{
|
| 585 |
+
"epoch": 2.7042074286168774,
|
| 586 |
+
"grad_norm": 2.6816985607147217,
|
| 587 |
+
"learning_rate": 0.00010989421776823351,
|
| 588 |
+
"loss": 0.1658,
|
| 589 |
+
"step": 34000
|
| 590 |
+
},
|
| 591 |
+
{
|
| 592 |
+
"epoch": 2.7439751849200666,
|
| 593 |
+
"grad_norm": 1.2923789024353027,
|
| 594 |
+
"learning_rate": 0.00010856862589146055,
|
| 595 |
+
"loss": 0.1588,
|
| 596 |
+
"step": 34500
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 2.783742941223256,
|
| 600 |
+
"grad_norm": 1.7398520708084106,
|
| 601 |
+
"learning_rate": 0.00010724568519844111,
|
| 602 |
+
"loss": 0.1609,
|
| 603 |
+
"step": 35000
|
| 604 |
+
},
|
| 605 |
+
{
|
| 606 |
+
"epoch": 2.800445398870596,
|
| 607 |
+
"eval_loss": 0.17199809849262238,
|
| 608 |
+
"eval_runtime": 45.1985,
|
| 609 |
+
"eval_samples_per_second": 11.062,
|
| 610 |
+
"eval_steps_per_second": 2.766,
|
| 611 |
+
"step": 35210
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 2.8235106975264457,
|
| 615 |
+
"grad_norm": 1.2712150812149048,
|
| 616 |
+
"learning_rate": 0.00010592009332166812,
|
| 617 |
+
"loss": 0.1539,
|
| 618 |
+
"step": 35500
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 2.863278453829635,
|
| 622 |
+
"grad_norm": 1.7284294366836548,
|
| 623 |
+
"learning_rate": 0.00010459450144489515,
|
| 624 |
+
"loss": 0.1587,
|
| 625 |
+
"step": 36000
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 2.9030462101328243,
|
| 629 |
+
"grad_norm": 1.1943145990371704,
|
| 630 |
+
"learning_rate": 0.00010326890956812216,
|
| 631 |
+
"loss": 0.1583,
|
| 632 |
+
"step": 36500
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 2.9428139664360136,
|
| 636 |
+
"grad_norm": 1.2150073051452637,
|
| 637 |
+
"learning_rate": 0.00010194862005885628,
|
| 638 |
+
"loss": 0.1526,
|
| 639 |
+
"step": 37000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 2.9825817227392033,
|
| 643 |
+
"grad_norm": 1.9882619380950928,
|
| 644 |
+
"learning_rate": 0.00010062302818208331,
|
| 645 |
+
"loss": 0.1571,
|
| 646 |
+
"step": 37500
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 3.000477213075638,
|
| 650 |
+
"eval_loss": 0.16704925894737244,
|
| 651 |
+
"eval_runtime": 45.457,
|
| 652 |
+
"eval_samples_per_second": 10.999,
|
| 653 |
+
"eval_steps_per_second": 2.75,
|
| 654 |
+
"step": 37725
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 3.0223494790423926,
|
| 658 |
+
"grad_norm": 1.0150736570358276,
|
| 659 |
+
"learning_rate": 9.929743630531032e-05,
|
| 660 |
+
"loss": 0.1381,
|
| 661 |
+
"step": 38000
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 3.062117235345582,
|
| 665 |
+
"grad_norm": 1.5414146184921265,
|
| 666 |
+
"learning_rate": 9.797184442853735e-05,
|
| 667 |
+
"loss": 0.1336,
|
| 668 |
+
"step": 38500
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 3.101884991648771,
|
| 672 |
+
"grad_norm": 1.8868834972381592,
|
| 673 |
+
"learning_rate": 9.664625255176437e-05,
|
| 674 |
+
"loss": 0.1421,
|
| 675 |
+
"step": 39000
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 3.1416527479519605,
|
| 679 |
+
"grad_norm": 1.9765745401382446,
|
| 680 |
+
"learning_rate": 9.532066067499139e-05,
|
| 681 |
+
"loss": 0.1371,
|
| 682 |
+
"step": 39500
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 3.1814205042551498,
|
| 686 |
+
"grad_norm": 0.8887368440628052,
|
| 687 |
+
"learning_rate": 9.399506879821841e-05,
|
| 688 |
+
"loss": 0.1301,
|
| 689 |
+
"step": 40000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 3.2005090272806807,
|
| 693 |
+
"eval_loss": 0.16935083270072937,
|
| 694 |
+
"eval_runtime": 45.1728,
|
| 695 |
+
"eval_samples_per_second": 11.069,
|
| 696 |
+
"eval_steps_per_second": 2.767,
|
| 697 |
+
"step": 40240
|
| 698 |
+
},
|
| 699 |
+
{
|
| 700 |
+
"epoch": 3.221188260558339,
|
| 701 |
+
"grad_norm": 1.5793101787567139,
|
| 702 |
+
"learning_rate": 9.266947692144543e-05,
|
| 703 |
+
"loss": 0.1377,
|
| 704 |
+
"step": 40500
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 3.260956016861529,
|
| 708 |
+
"grad_norm": 0.8649873733520508,
|
| 709 |
+
"learning_rate": 9.134653622842599e-05,
|
| 710 |
+
"loss": 0.1374,
|
| 711 |
+
"step": 41000
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 3.300723773164718,
|
| 715 |
+
"grad_norm": 1.7010825872421265,
|
| 716 |
+
"learning_rate": 9.002094435165301e-05,
|
| 717 |
+
"loss": 0.1371,
|
| 718 |
+
"step": 41500
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 3.3404915294679074,
|
| 722 |
+
"grad_norm": 2.9131572246551514,
|
| 723 |
+
"learning_rate": 8.869535247488003e-05,
|
| 724 |
+
"loss": 0.1371,
|
| 725 |
+
"step": 42000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 3.3802592857710967,
|
| 729 |
+
"grad_norm": 1.7646764516830444,
|
| 730 |
+
"learning_rate": 8.736976059810706e-05,
|
| 731 |
+
"loss": 0.1362,
|
| 732 |
+
"step": 42500
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 3.400540841485723,
|
| 736 |
+
"eval_loss": 0.16898168623447418,
|
| 737 |
+
"eval_runtime": 45.1702,
|
| 738 |
+
"eval_samples_per_second": 11.069,
|
| 739 |
+
"eval_steps_per_second": 2.767,
|
| 740 |
+
"step": 42755
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 3.420027042074286,
|
| 744 |
+
"grad_norm": 1.3960251808166504,
|
| 745 |
+
"learning_rate": 8.604416872133408e-05,
|
| 746 |
+
"loss": 0.1387,
|
| 747 |
+
"step": 43000
|
| 748 |
+
},
|
| 749 |
+
{
|
| 750 |
+
"epoch": 3.4597947983774757,
|
| 751 |
+
"grad_norm": 1.9521344900131226,
|
| 752 |
+
"learning_rate": 8.47185768445611e-05,
|
| 753 |
+
"loss": 0.1371,
|
| 754 |
+
"step": 43500
|
| 755 |
+
},
|
| 756 |
+
{
|
| 757 |
+
"epoch": 3.499562554680665,
|
| 758 |
+
"grad_norm": 1.8496028184890747,
|
| 759 |
+
"learning_rate": 8.339298496778812e-05,
|
| 760 |
+
"loss": 0.1366,
|
| 761 |
+
"step": 44000
|
| 762 |
+
},
|
| 763 |
+
{
|
| 764 |
+
"epoch": 3.5393303109838543,
|
| 765 |
+
"grad_norm": 1.118754267692566,
|
| 766 |
+
"learning_rate": 8.207004427476868e-05,
|
| 767 |
+
"loss": 0.1352,
|
| 768 |
+
"step": 44500
|
| 769 |
+
},
|
| 770 |
+
{
|
| 771 |
+
"epoch": 3.5790980672870436,
|
| 772 |
+
"grad_norm": 1.3573617935180664,
|
| 773 |
+
"learning_rate": 8.07444523979957e-05,
|
| 774 |
+
"loss": 0.1384,
|
| 775 |
+
"step": 45000
|
| 776 |
+
},
|
| 777 |
+
{
|
| 778 |
+
"epoch": 3.600572655690766,
|
| 779 |
+
"eval_loss": 0.1594919115304947,
|
| 780 |
+
"eval_runtime": 45.1624,
|
| 781 |
+
"eval_samples_per_second": 11.071,
|
| 782 |
+
"eval_steps_per_second": 2.768,
|
| 783 |
+
"step": 45270
|
| 784 |
+
},
|
| 785 |
+
{
|
| 786 |
+
"epoch": 3.618865823590233,
|
| 787 |
+
"grad_norm": 1.723669171333313,
|
| 788 |
+
"learning_rate": 7.941886052122272e-05,
|
| 789 |
+
"loss": 0.135,
|
| 790 |
+
"step": 45500
|
| 791 |
+
},
|
| 792 |
+
{
|
| 793 |
+
"epoch": 3.658633579893422,
|
| 794 |
+
"grad_norm": 1.381091594696045,
|
| 795 |
+
"learning_rate": 7.809326864444974e-05,
|
| 796 |
+
"loss": 0.1367,
|
| 797 |
+
"step": 46000
|
| 798 |
+
},
|
| 799 |
+
{
|
| 800 |
+
"epoch": 3.698401336196612,
|
| 801 |
+
"grad_norm": 1.2882030010223389,
|
| 802 |
+
"learning_rate": 7.676767676767676e-05,
|
| 803 |
+
"loss": 0.1321,
|
| 804 |
+
"step": 46500
|
| 805 |
+
},
|
| 806 |
+
{
|
| 807 |
+
"epoch": 3.7381690924998012,
|
| 808 |
+
"grad_norm": 0.6370302438735962,
|
| 809 |
+
"learning_rate": 7.544473607465734e-05,
|
| 810 |
+
"loss": 0.1315,
|
| 811 |
+
"step": 47000
|
| 812 |
+
},
|
| 813 |
+
{
|
| 814 |
+
"epoch": 3.7779368488029905,
|
| 815 |
+
"grad_norm": 1.4381242990493774,
|
| 816 |
+
"learning_rate": 7.411914419788436e-05,
|
| 817 |
+
"loss": 0.1296,
|
| 818 |
+
"step": 47500
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 3.8006044698958084,
|
| 822 |
+
"eval_loss": 0.15816625952720642,
|
| 823 |
+
"eval_runtime": 45.2488,
|
| 824 |
+
"eval_samples_per_second": 11.05,
|
| 825 |
+
"eval_steps_per_second": 2.763,
|
| 826 |
+
"step": 47785
|
| 827 |
+
},
|
| 828 |
+
{
|
| 829 |
+
"epoch": 3.81770460510618,
|
| 830 |
+
"grad_norm": 1.8064157962799072,
|
| 831 |
+
"learning_rate": 7.279355232111138e-05,
|
| 832 |
+
"loss": 0.1313,
|
| 833 |
+
"step": 48000
|
| 834 |
+
},
|
| 835 |
+
{
|
| 836 |
+
"epoch": 3.8574723614093696,
|
| 837 |
+
"grad_norm": 1.9430205821990967,
|
| 838 |
+
"learning_rate": 7.14679604443384e-05,
|
| 839 |
+
"loss": 0.1358,
|
| 840 |
+
"step": 48500
|
| 841 |
+
},
|
| 842 |
+
{
|
| 843 |
+
"epoch": 3.897240117712559,
|
| 844 |
+
"grad_norm": 2.3239598274230957,
|
| 845 |
+
"learning_rate": 7.014501975131896e-05,
|
| 846 |
+
"loss": 0.1339,
|
| 847 |
+
"step": 49000
|
| 848 |
+
},
|
| 849 |
+
{
|
| 850 |
+
"epoch": 3.937007874015748,
|
| 851 |
+
"grad_norm": 1.2267085313796997,
|
| 852 |
+
"learning_rate": 6.881942787454598e-05,
|
| 853 |
+
"loss": 0.1321,
|
| 854 |
+
"step": 49500
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 3.9767756303189374,
|
| 858 |
+
"grad_norm": 0.8892061114311218,
|
| 859 |
+
"learning_rate": 6.749383599777301e-05,
|
| 860 |
+
"loss": 0.1332,
|
| 861 |
+
"step": 50000
|
| 862 |
+
},
|
| 863 |
+
{
|
| 864 |
+
"epoch": 4.000636284100851,
|
| 865 |
+
"eval_loss": 0.15572361648082733,
|
| 866 |
+
"eval_runtime": 45.254,
|
| 867 |
+
"eval_samples_per_second": 11.049,
|
| 868 |
+
"eval_steps_per_second": 2.762,
|
| 869 |
+
"step": 50300
|
| 870 |
+
},
|
| 871 |
+
{
|
| 872 |
+
"epoch": 4.016543386622127,
|
| 873 |
+
"grad_norm": 1.5816055536270142,
|
| 874 |
+
"learning_rate": 6.616824412100003e-05,
|
| 875 |
+
"loss": 0.1207,
|
| 876 |
+
"step": 50500
|
| 877 |
+
},
|
| 878 |
+
{
|
| 879 |
+
"epoch": 4.056311142925316,
|
| 880 |
+
"grad_norm": 1.8249752521514893,
|
| 881 |
+
"learning_rate": 6.484265224422705e-05,
|
| 882 |
+
"loss": 0.1135,
|
| 883 |
+
"step": 51000
|
| 884 |
+
},
|
| 885 |
+
{
|
| 886 |
+
"epoch": 4.096078899228505,
|
| 887 |
+
"grad_norm": 2.009570837020874,
|
| 888 |
+
"learning_rate": 6.351706036745407e-05,
|
| 889 |
+
"loss": 0.1101,
|
| 890 |
+
"step": 51500
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 4.135846655531695,
|
| 894 |
+
"grad_norm": 2.322263240814209,
|
| 895 |
+
"learning_rate": 6.219146849068109e-05,
|
| 896 |
+
"loss": 0.1123,
|
| 897 |
+
"step": 52000
|
| 898 |
+
},
|
| 899 |
+
{
|
| 900 |
+
"epoch": 4.175614411834884,
|
| 901 |
+
"grad_norm": 0.8762745261192322,
|
| 902 |
+
"learning_rate": 6.086587661390811e-05,
|
| 903 |
+
"loss": 0.1122,
|
| 904 |
+
"step": 52500
|
| 905 |
+
},
|
| 906 |
+
{
|
| 907 |
+
"epoch": 4.2006680983058935,
|
| 908 |
+
"eval_loss": 0.15810129046440125,
|
| 909 |
+
"eval_runtime": 45.1569,
|
| 910 |
+
"eval_samples_per_second": 11.072,
|
| 911 |
+
"eval_steps_per_second": 2.768,
|
| 912 |
+
"step": 52815
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 4.215382168138074,
|
| 916 |
+
"grad_norm": 2.09260892868042,
|
| 917 |
+
"learning_rate": 5.954028473713513e-05,
|
| 918 |
+
"loss": 0.1115,
|
| 919 |
+
"step": 53000
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 4.255149924441263,
|
| 923 |
+
"grad_norm": 1.0858501195907593,
|
| 924 |
+
"learning_rate": 5.821469286036215e-05,
|
| 925 |
+
"loss": 0.1152,
|
| 926 |
+
"step": 53500
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 4.294917680744453,
|
| 930 |
+
"grad_norm": 1.889620065689087,
|
| 931 |
+
"learning_rate": 5.688910098358917e-05,
|
| 932 |
+
"loss": 0.1118,
|
| 933 |
+
"step": 54000
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 4.334685437047642,
|
| 937 |
+
"grad_norm": 2.019217014312744,
|
| 938 |
+
"learning_rate": 5.556350910681619e-05,
|
| 939 |
+
"loss": 0.1157,
|
| 940 |
+
"step": 54500
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 4.374453193350831,
|
| 944 |
+
"grad_norm": 2.212111473083496,
|
| 945 |
+
"learning_rate": 5.424056841379677e-05,
|
| 946 |
+
"loss": 0.1116,
|
| 947 |
+
"step": 55000
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 4.4006999125109365,
|
| 951 |
+
"eval_loss": 0.15376093983650208,
|
| 952 |
+
"eval_runtime": 45.1499,
|
| 953 |
+
"eval_samples_per_second": 11.074,
|
| 954 |
+
"eval_steps_per_second": 2.769,
|
| 955 |
+
"step": 55330
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 4.414220949654021,
|
| 959 |
+
"grad_norm": 1.7018357515335083,
|
| 960 |
+
"learning_rate": 5.291762772077733e-05,
|
| 961 |
+
"loss": 0.117,
|
| 962 |
+
"step": 55500
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 4.45398870595721,
|
| 966 |
+
"grad_norm": 1.5110944509506226,
|
| 967 |
+
"learning_rate": 5.159203584400435e-05,
|
| 968 |
+
"loss": 0.1165,
|
| 969 |
+
"step": 56000
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 4.493756462260399,
|
| 973 |
+
"grad_norm": 2.047557830810547,
|
| 974 |
+
"learning_rate": 5.026644396723137e-05,
|
| 975 |
+
"loss": 0.1125,
|
| 976 |
+
"step": 56500
|
| 977 |
+
},
|
| 978 |
+
{
|
| 979 |
+
"epoch": 4.5335242185635884,
|
| 980 |
+
"grad_norm": 0.9929693341255188,
|
| 981 |
+
"learning_rate": 4.8940852090458395e-05,
|
| 982 |
+
"loss": 0.1049,
|
| 983 |
+
"step": 57000
|
| 984 |
+
},
|
| 985 |
+
{
|
| 986 |
+
"epoch": 4.573291974866778,
|
| 987 |
+
"grad_norm": 3.9604108333587646,
|
| 988 |
+
"learning_rate": 4.761791139743896e-05,
|
| 989 |
+
"loss": 0.1093,
|
| 990 |
+
"step": 57500
|
| 991 |
+
},
|
| 992 |
+
{
|
| 993 |
+
"epoch": 4.600731726715979,
|
| 994 |
+
"eval_loss": 0.15253813564777374,
|
| 995 |
+
"eval_runtime": 45.2381,
|
| 996 |
+
"eval_samples_per_second": 11.053,
|
| 997 |
+
"eval_steps_per_second": 2.763,
|
| 998 |
+
"step": 57845
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 4.613059731169967,
|
| 1002 |
+
"grad_norm": 1.9711782932281494,
|
| 1003 |
+
"learning_rate": 4.629497070441953e-05,
|
| 1004 |
+
"loss": 0.1104,
|
| 1005 |
+
"step": 58000
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 4.652827487473157,
|
| 1009 |
+
"grad_norm": 1.9756906032562256,
|
| 1010 |
+
"learning_rate": 4.496937882764655e-05,
|
| 1011 |
+
"loss": 0.1095,
|
| 1012 |
+
"step": 58500
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 4.6925952437763465,
|
| 1016 |
+
"grad_norm": 2.463798761367798,
|
| 1017 |
+
"learning_rate": 4.364378695087357e-05,
|
| 1018 |
+
"loss": 0.11,
|
| 1019 |
+
"step": 59000
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 4.732363000079536,
|
| 1023 |
+
"grad_norm": 6.79127836227417,
|
| 1024 |
+
"learning_rate": 4.231819507410059e-05,
|
| 1025 |
+
"loss": 0.1066,
|
| 1026 |
+
"step": 59500
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 4.772130756382725,
|
| 1030 |
+
"grad_norm": 1.2049651145935059,
|
| 1031 |
+
"learning_rate": 4.0992603197327614e-05,
|
| 1032 |
+
"loss": 0.1082,
|
| 1033 |
+
"step": 60000
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 4.800763540921022,
|
| 1037 |
+
"eval_loss": 0.15342682600021362,
|
| 1038 |
+
"eval_runtime": 45.3206,
|
| 1039 |
+
"eval_samples_per_second": 11.032,
|
| 1040 |
+
"eval_steps_per_second": 2.758,
|
| 1041 |
+
"step": 60360
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 4.811898512685914,
|
| 1045 |
+
"grad_norm": 1.2515356540679932,
|
| 1046 |
+
"learning_rate": 3.966701132055463e-05,
|
| 1047 |
+
"loss": 0.1147,
|
| 1048 |
+
"step": 60500
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 4.851666268989104,
|
| 1052 |
+
"grad_norm": 1.2210029363632202,
|
| 1053 |
+
"learning_rate": 3.834141944378165e-05,
|
| 1054 |
+
"loss": 0.1124,
|
| 1055 |
+
"step": 61000
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 4.891434025292293,
|
| 1059 |
+
"grad_norm": 3.2822036743164062,
|
| 1060 |
+
"learning_rate": 3.701582756700867e-05,
|
| 1061 |
+
"loss": 0.1107,
|
| 1062 |
+
"step": 61500
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 4.931201781595482,
|
| 1066 |
+
"grad_norm": 1.3503117561340332,
|
| 1067 |
+
"learning_rate": 3.569023569023569e-05,
|
| 1068 |
+
"loss": 0.1062,
|
| 1069 |
+
"step": 62000
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 4.970969537898672,
|
| 1073 |
+
"grad_norm": 2.045919418334961,
|
| 1074 |
+
"learning_rate": 3.436464381346271e-05,
|
| 1075 |
+
"loss": 0.1056,
|
| 1076 |
+
"step": 62500
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 5.000795355126064,
|
| 1080 |
+
"eval_loss": 0.15002138912677765,
|
| 1081 |
+
"eval_runtime": 45.147,
|
| 1082 |
+
"eval_samples_per_second": 11.075,
|
| 1083 |
+
"eval_steps_per_second": 2.769,
|
| 1084 |
+
"step": 62875
|
| 1085 |
+
},
|
| 1086 |
+
{
|
| 1087 |
+
"epoch": 5.010737294201861,
|
| 1088 |
+
"grad_norm": 1.0435826778411865,
|
| 1089 |
+
"learning_rate": 3.303905193668974e-05,
|
| 1090 |
+
"loss": 0.1064,
|
| 1091 |
+
"step": 63000
|
| 1092 |
+
},
|
| 1093 |
+
{
|
| 1094 |
+
"epoch": 5.05050505050505,
|
| 1095 |
+
"grad_norm": 2.05426287651062,
|
| 1096 |
+
"learning_rate": 3.171346005991676e-05,
|
| 1097 |
+
"loss": 0.0916,
|
| 1098 |
+
"step": 63500
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 5.0902728068082395,
|
| 1102 |
+
"grad_norm": 3.189483880996704,
|
| 1103 |
+
"learning_rate": 3.0390519366897317e-05,
|
| 1104 |
+
"loss": 0.0913,
|
| 1105 |
+
"step": 64000
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 5.13004056311143,
|
| 1109 |
+
"grad_norm": 1.1413475275039673,
|
| 1110 |
+
"learning_rate": 2.9064927490124343e-05,
|
| 1111 |
+
"loss": 0.0923,
|
| 1112 |
+
"step": 64500
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 5.169808319414619,
|
| 1116 |
+
"grad_norm": 0.778860330581665,
|
| 1117 |
+
"learning_rate": 2.7739335613351363e-05,
|
| 1118 |
+
"loss": 0.0904,
|
| 1119 |
+
"step": 65000
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 5.200827169331106,
|
| 1123 |
+
"eval_loss": 0.1528937965631485,
|
| 1124 |
+
"eval_runtime": 45.1301,
|
| 1125 |
+
"eval_samples_per_second": 11.079,
|
| 1126 |
+
"eval_steps_per_second": 2.77,
|
| 1127 |
+
"step": 65390
|
| 1128 |
+
},
|
| 1129 |
+
{
|
| 1130 |
+
"epoch": 5.209576075717808,
|
| 1131 |
+
"grad_norm": 1.8860127925872803,
|
| 1132 |
+
"learning_rate": 2.6413743736578382e-05,
|
| 1133 |
+
"loss": 0.0936,
|
| 1134 |
+
"step": 65500
|
| 1135 |
+
},
|
| 1136 |
+
{
|
| 1137 |
+
"epoch": 5.2493438320209975,
|
| 1138 |
+
"grad_norm": 1.6669965982437134,
|
| 1139 |
+
"learning_rate": 2.5093454227312496e-05,
|
| 1140 |
+
"loss": 0.095,
|
| 1141 |
+
"step": 66000
|
| 1142 |
+
},
|
| 1143 |
+
{
|
| 1144 |
+
"epoch": 5.289111588324187,
|
| 1145 |
+
"grad_norm": 2.063001871109009,
|
| 1146 |
+
"learning_rate": 2.3767862350539516e-05,
|
| 1147 |
+
"loss": 0.0915,
|
| 1148 |
+
"step": 66500
|
| 1149 |
+
},
|
| 1150 |
+
{
|
| 1151 |
+
"epoch": 5.328879344627376,
|
| 1152 |
+
"grad_norm": 1.8162118196487427,
|
| 1153 |
+
"learning_rate": 2.244227047376654e-05,
|
| 1154 |
+
"loss": 0.09,
|
| 1155 |
+
"step": 67000
|
| 1156 |
+
},
|
| 1157 |
+
{
|
| 1158 |
+
"epoch": 5.368647100930565,
|
| 1159 |
+
"grad_norm": 1.5958954095840454,
|
| 1160 |
+
"learning_rate": 2.1116678596993558e-05,
|
| 1161 |
+
"loss": 0.0901,
|
| 1162 |
+
"step": 67500
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 5.400858983536149,
|
| 1166 |
+
"eval_loss": 0.15164533257484436,
|
| 1167 |
+
"eval_runtime": 45.423,
|
| 1168 |
+
"eval_samples_per_second": 11.008,
|
| 1169 |
+
"eval_steps_per_second": 2.752,
|
| 1170 |
+
"step": 67905
|
| 1171 |
+
},
|
| 1172 |
+
{
|
| 1173 |
+
"epoch": 5.408414857233755,
|
| 1174 |
+
"grad_norm": 0.7466435432434082,
|
| 1175 |
+
"learning_rate": 1.9793737903974125e-05,
|
| 1176 |
+
"loss": 0.0919,
|
| 1177 |
+
"step": 68000
|
| 1178 |
+
},
|
| 1179 |
+
{
|
| 1180 |
+
"epoch": 5.448182613536944,
|
| 1181 |
+
"grad_norm": 1.3805254697799683,
|
| 1182 |
+
"learning_rate": 1.8468146027201148e-05,
|
| 1183 |
+
"loss": 0.0896,
|
| 1184 |
+
"step": 68500
|
| 1185 |
+
},
|
| 1186 |
+
{
|
| 1187 |
+
"epoch": 5.487950369840133,
|
| 1188 |
+
"grad_norm": 1.2593927383422852,
|
| 1189 |
+
"learning_rate": 1.7142554150428167e-05,
|
| 1190 |
+
"loss": 0.0914,
|
| 1191 |
+
"step": 69000
|
| 1192 |
+
},
|
| 1193 |
+
{
|
| 1194 |
+
"epoch": 5.5277181261433235,
|
| 1195 |
+
"grad_norm": 1.3591930866241455,
|
| 1196 |
+
"learning_rate": 1.5816962273655187e-05,
|
| 1197 |
+
"loss": 0.0892,
|
| 1198 |
+
"step": 69500
|
| 1199 |
+
},
|
| 1200 |
+
{
|
| 1201 |
+
"epoch": 5.567485882446512,
|
| 1202 |
+
"grad_norm": 1.918165922164917,
|
| 1203 |
+
"learning_rate": 1.4491370396882208e-05,
|
| 1204 |
+
"loss": 0.0893,
|
| 1205 |
+
"step": 70000
|
| 1206 |
+
},
|
| 1207 |
+
{
|
| 1208 |
+
"epoch": 5.600890797741192,
|
| 1209 |
+
"eval_loss": 0.14885325729846954,
|
| 1210 |
+
"eval_runtime": 45.4575,
|
| 1211 |
+
"eval_samples_per_second": 10.999,
|
| 1212 |
+
"eval_steps_per_second": 2.75,
|
| 1213 |
+
"step": 70420
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 5.607253638749702,
|
| 1217 |
+
"grad_norm": 1.2859373092651367,
|
| 1218 |
+
"learning_rate": 1.3168429703862775e-05,
|
| 1219 |
+
"loss": 0.0863,
|
| 1220 |
+
"step": 70500
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 5.647021395052891,
|
| 1224 |
+
"grad_norm": 1.445563554763794,
|
| 1225 |
+
"learning_rate": 1.1842837827089796e-05,
|
| 1226 |
+
"loss": 0.0888,
|
| 1227 |
+
"step": 71000
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 5.686789151356081,
|
| 1231 |
+
"grad_norm": 1.5366461277008057,
|
| 1232 |
+
"learning_rate": 1.0517245950316817e-05,
|
| 1233 |
+
"loss": 0.0894,
|
| 1234 |
+
"step": 71500
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 5.72655690765927,
|
| 1238 |
+
"grad_norm": 1.1700299978256226,
|
| 1239 |
+
"learning_rate": 9.191654073543837e-06,
|
| 1240 |
+
"loss": 0.0968,
|
| 1241 |
+
"step": 72000
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 5.766324663962459,
|
| 1245 |
+
"grad_norm": 1.7285205125808716,
|
| 1246 |
+
"learning_rate": 7.868713380524405e-06,
|
| 1247 |
+
"loss": 0.0928,
|
| 1248 |
+
"step": 72500
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 5.800922611946234,
|
| 1252 |
+
"eval_loss": 0.15042120218276978,
|
| 1253 |
+
"eval_runtime": 45.1582,
|
| 1254 |
+
"eval_samples_per_second": 11.072,
|
| 1255 |
+
"eval_steps_per_second": 2.768,
|
| 1256 |
+
"step": 72935
|
| 1257 |
+
},
|
| 1258 |
+
{
|
| 1259 |
+
"epoch": 5.8060924202656485,
|
| 1260 |
+
"grad_norm": 1.7231446504592896,
|
| 1261 |
+
"learning_rate": 6.543121503751426e-06,
|
| 1262 |
+
"loss": 0.0907,
|
| 1263 |
+
"step": 73000
|
| 1264 |
+
},
|
| 1265 |
+
{
|
| 1266 |
+
"epoch": 5.845860176568838,
|
| 1267 |
+
"grad_norm": 0.8501458764076233,
|
| 1268 |
+
"learning_rate": 5.217529626978446e-06,
|
| 1269 |
+
"loss": 0.0906,
|
| 1270 |
+
"step": 73500
|
| 1271 |
+
},
|
| 1272 |
+
{
|
| 1273 |
+
"epoch": 5.885627932872027,
|
| 1274 |
+
"grad_norm": 1.1770985126495361,
|
| 1275 |
+
"learning_rate": 3.891937750205467e-06,
|
| 1276 |
+
"loss": 0.0919,
|
| 1277 |
+
"step": 74000
|
| 1278 |
+
},
|
| 1279 |
+
{
|
| 1280 |
+
"epoch": 5.925395689175216,
|
| 1281 |
+
"grad_norm": 2.067455530166626,
|
| 1282 |
+
"learning_rate": 2.5663458734324876e-06,
|
| 1283 |
+
"loss": 0.0873,
|
| 1284 |
+
"step": 74500
|
| 1285 |
+
},
|
| 1286 |
+
{
|
| 1287 |
+
"epoch": 5.965163445478407,
|
| 1288 |
+
"grad_norm": 2.9192771911621094,
|
| 1289 |
+
"learning_rate": 1.2407539966595084e-06,
|
| 1290 |
+
"loss": 0.0887,
|
| 1291 |
+
"step": 75000
|
| 1292 |
+
},
|
| 1293 |
+
{
|
| 1294 |
+
"epoch": 6.0,
|
| 1295 |
+
"step": 75438,
|
| 1296 |
+
"total_flos": 2.0219589895971348e+18,
|
| 1297 |
+
"train_loss": 0.16557818883334932,
|
| 1298 |
+
"train_runtime": 122761.9475,
|
| 1299 |
+
"train_samples_per_second": 2.458,
|
| 1300 |
+
"train_steps_per_second": 0.615
|
| 1301 |
+
}
|
| 1302 |
+
],
|
| 1303 |
+
"logging_steps": 500,
|
| 1304 |
+
"max_steps": 75438,
|
| 1305 |
+
"num_input_tokens_seen": 0,
|
| 1306 |
+
"num_train_epochs": 6,
|
| 1307 |
+
"save_steps": 500,
|
| 1308 |
+
"stateful_callbacks": {
|
| 1309 |
+
"TrainerControl": {
|
| 1310 |
+
"args": {
|
| 1311 |
+
"should_epoch_stop": false,
|
| 1312 |
+
"should_evaluate": false,
|
| 1313 |
+
"should_log": false,
|
| 1314 |
+
"should_save": true,
|
| 1315 |
+
"should_training_stop": true
|
| 1316 |
+
},
|
| 1317 |
+
"attributes": {}
|
| 1318 |
+
}
|
| 1319 |
+
},
|
| 1320 |
+
"total_flos": 2.0219589895971348e+18,
|
| 1321 |
+
"train_batch_size": 4,
|
| 1322 |
+
"trial_name": null,
|
| 1323 |
+
"trial_params": null
|
| 1324 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:de9993fdf9cc959da3d3c449dfb8030489386d8c59084b5eb3876618e28cd047
|
| 3 |
+
size 5496
|