End of training
Browse files- README.md +89 -0
- adapter_config.json +32 -0
- adapter_model.safetensors +3 -0
- trainer_state.json +2381 -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-processed-task1_min_symbols_lemma_object_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-processed-task1_min_symbols_lemma_object_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: 0.2167
|
| 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 |
+
| 0.4548 | 0.2000 | 5030 | 0.4143 |
|
| 53 |
+
| 0.3861 | 0.4001 | 10060 | 0.3652 |
|
| 54 |
+
| 0.3504 | 0.6001 | 15090 | 0.3359 |
|
| 55 |
+
| 0.3352 | 0.8001 | 20120 | 0.3129 |
|
| 56 |
+
| 0.3186 | 1.0002 | 25150 | 0.3011 |
|
| 57 |
+
| 0.2763 | 1.2002 | 30180 | 0.2905 |
|
| 58 |
+
| 0.2711 | 1.4002 | 35210 | 0.2776 |
|
| 59 |
+
| 0.2579 | 1.6003 | 40240 | 0.2743 |
|
| 60 |
+
| 0.2484 | 1.8003 | 45270 | 0.2600 |
|
| 61 |
+
| 0.263 | 2.0003 | 50300 | 0.2602 |
|
| 62 |
+
| 0.2108 | 2.2003 | 55330 | 0.2580 |
|
| 63 |
+
| 0.2026 | 2.4004 | 60360 | 0.2526 |
|
| 64 |
+
| 0.2228 | 2.6004 | 65390 | 0.2473 |
|
| 65 |
+
| 0.2187 | 2.8004 | 70420 | 0.2387 |
|
| 66 |
+
| 0.2053 | 3.0005 | 75450 | 0.2345 |
|
| 67 |
+
| 0.1761 | 3.2005 | 80480 | 0.2379 |
|
| 68 |
+
| 0.1773 | 3.4005 | 85510 | 0.2323 |
|
| 69 |
+
| 0.1689 | 3.6006 | 90540 | 0.2265 |
|
| 70 |
+
| 0.1707 | 3.8006 | 95570 | 0.2256 |
|
| 71 |
+
| 0.1764 | 4.0006 | 100600 | 0.2197 |
|
| 72 |
+
| 0.1377 | 4.2007 | 105630 | 0.2262 |
|
| 73 |
+
| 0.1485 | 4.4007 | 110660 | 0.2230 |
|
| 74 |
+
| 0.1464 | 4.6007 | 115690 | 0.2161 |
|
| 75 |
+
| 0.1521 | 4.8008 | 120720 | 0.2178 |
|
| 76 |
+
| 0.1457 | 5.0008 | 125750 | 0.2144 |
|
| 77 |
+
| 0.1146 | 5.2008 | 130780 | 0.2221 |
|
| 78 |
+
| 0.1187 | 5.4009 | 135810 | 0.2204 |
|
| 79 |
+
| 0.1041 | 5.6009 | 140840 | 0.2201 |
|
| 80 |
+
| 0.1171 | 5.8009 | 145870 | 0.2167 |
|
| 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 sha256:1a00063faa09dea22ff61eb26f325740133a9e27444dc8536846d4af0e32a97c
|
| 3 |
+
size 531035104
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2381 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 6.0,
|
| 5 |
+
"eval_steps": 5030,
|
| 6 |
+
"global_step": 150876,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.019883878151594685,
|
| 13 |
+
"grad_norm": 3.941704273223877,
|
| 14 |
+
"learning_rate": 0.00019947206451955218,
|
| 15 |
+
"loss": 0.7704,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.03976775630318937,
|
| 20 |
+
"grad_norm": 3.321659564971924,
|
| 21 |
+
"learning_rate": 0.00019880882899135142,
|
| 22 |
+
"loss": 0.5987,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.05965163445478406,
|
| 27 |
+
"grad_norm": 2.6819937229156494,
|
| 28 |
+
"learning_rate": 0.00019814559346315066,
|
| 29 |
+
"loss": 0.552,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.07953551260637874,
|
| 34 |
+
"grad_norm": 1.794763445854187,
|
| 35 |
+
"learning_rate": 0.00019748235793494987,
|
| 36 |
+
"loss": 0.5215,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.09941939075797343,
|
| 41 |
+
"grad_norm": 1.0604099035263062,
|
| 42 |
+
"learning_rate": 0.00019681912240674908,
|
| 43 |
+
"loss": 0.5051,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.11930326890956812,
|
| 48 |
+
"grad_norm": 1.685113787651062,
|
| 49 |
+
"learning_rate": 0.00019615721334960472,
|
| 50 |
+
"loss": 0.4871,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.1391871470611628,
|
| 55 |
+
"grad_norm": 2.0144481658935547,
|
| 56 |
+
"learning_rate": 0.00019549397782140396,
|
| 57 |
+
"loss": 0.4651,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.15907102521275748,
|
| 62 |
+
"grad_norm": 2.2157390117645264,
|
| 63 |
+
"learning_rate": 0.00019483074229320317,
|
| 64 |
+
"loss": 0.435,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.17895490336435219,
|
| 69 |
+
"grad_norm": 2.2920103073120117,
|
| 70 |
+
"learning_rate": 0.00019416750676500238,
|
| 71 |
+
"loss": 0.4421,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.19883878151594686,
|
| 76 |
+
"grad_norm": 1.9781739711761475,
|
| 77 |
+
"learning_rate": 0.00019350559770785805,
|
| 78 |
+
"loss": 0.4548,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.20003181420504254,
|
| 83 |
+
"eval_loss": 0.4143450856208801,
|
| 84 |
+
"eval_runtime": 32.7843,
|
| 85 |
+
"eval_samples_per_second": 15.251,
|
| 86 |
+
"eval_steps_per_second": 7.626,
|
| 87 |
+
"step": 5030
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.21872265966754156,
|
| 91 |
+
"grad_norm": 3.0981860160827637,
|
| 92 |
+
"learning_rate": 0.00019284236217965726,
|
| 93 |
+
"loss": 0.4301,
|
| 94 |
+
"step": 5500
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.23860653781913624,
|
| 98 |
+
"grad_norm": 2.0863499641418457,
|
| 99 |
+
"learning_rate": 0.00019217912665145647,
|
| 100 |
+
"loss": 0.4222,
|
| 101 |
+
"step": 6000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.2584904159707309,
|
| 105 |
+
"grad_norm": 1.4780126810073853,
|
| 106 |
+
"learning_rate": 0.0001915158911232557,
|
| 107 |
+
"loss": 0.3854,
|
| 108 |
+
"step": 6500
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.2783742941223256,
|
| 112 |
+
"grad_norm": 1.96503484249115,
|
| 113 |
+
"learning_rate": 0.00019085398206611135,
|
| 114 |
+
"loss": 0.4078,
|
| 115 |
+
"step": 7000
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.2982581722739203,
|
| 119 |
+
"grad_norm": 1.284993290901184,
|
| 120 |
+
"learning_rate": 0.00019019074653791056,
|
| 121 |
+
"loss": 0.4056,
|
| 122 |
+
"step": 7500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.31814205042551497,
|
| 126 |
+
"grad_norm": 1.9425936937332153,
|
| 127 |
+
"learning_rate": 0.00018952751100970977,
|
| 128 |
+
"loss": 0.3963,
|
| 129 |
+
"step": 8000
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.33802592857710967,
|
| 133 |
+
"grad_norm": 1.3688462972640991,
|
| 134 |
+
"learning_rate": 0.000188864275481509,
|
| 135 |
+
"loss": 0.3997,
|
| 136 |
+
"step": 8500
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.35790980672870437,
|
| 140 |
+
"grad_norm": 2.033297538757324,
|
| 141 |
+
"learning_rate": 0.00018820103995330822,
|
| 142 |
+
"loss": 0.3976,
|
| 143 |
+
"step": 9000
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.3777936848802991,
|
| 147 |
+
"grad_norm": 1.5158119201660156,
|
| 148 |
+
"learning_rate": 0.00018753780442510743,
|
| 149 |
+
"loss": 0.3841,
|
| 150 |
+
"step": 9500
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.3976775630318937,
|
| 154 |
+
"grad_norm": 5.128476619720459,
|
| 155 |
+
"learning_rate": 0.00018687456889690667,
|
| 156 |
+
"loss": 0.3861,
|
| 157 |
+
"step": 10000
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.4000636284100851,
|
| 161 |
+
"eval_loss": 0.36515533924102783,
|
| 162 |
+
"eval_runtime": 34.0872,
|
| 163 |
+
"eval_samples_per_second": 14.668,
|
| 164 |
+
"eval_steps_per_second": 7.334,
|
| 165 |
+
"step": 10060
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.4175614411834884,
|
| 169 |
+
"grad_norm": 2.8938169479370117,
|
| 170 |
+
"learning_rate": 0.00018621133336870591,
|
| 171 |
+
"loss": 0.3658,
|
| 172 |
+
"step": 10500
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.4374453193350831,
|
| 176 |
+
"grad_norm": 3.732710599899292,
|
| 177 |
+
"learning_rate": 0.00018554809784050513,
|
| 178 |
+
"loss": 0.3837,
|
| 179 |
+
"step": 11000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.4573291974866778,
|
| 183 |
+
"grad_norm": 1.6790924072265625,
|
| 184 |
+
"learning_rate": 0.00018488486231230434,
|
| 185 |
+
"loss": 0.3788,
|
| 186 |
+
"step": 11500
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.4772130756382725,
|
| 190 |
+
"grad_norm": 3.105046272277832,
|
| 191 |
+
"learning_rate": 0.00018422162678410358,
|
| 192 |
+
"loss": 0.3646,
|
| 193 |
+
"step": 12000
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.4970969537898672,
|
| 197 |
+
"grad_norm": 1.513545036315918,
|
| 198 |
+
"learning_rate": 0.00018355971772695921,
|
| 199 |
+
"loss": 0.3736,
|
| 200 |
+
"step": 12500
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.5169808319414618,
|
| 204 |
+
"grad_norm": 1.7431304454803467,
|
| 205 |
+
"learning_rate": 0.00018289648219875843,
|
| 206 |
+
"loss": 0.3715,
|
| 207 |
+
"step": 13000
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.5368647100930566,
|
| 211 |
+
"grad_norm": 1.9994251728057861,
|
| 212 |
+
"learning_rate": 0.00018223324667055767,
|
| 213 |
+
"loss": 0.3446,
|
| 214 |
+
"step": 13500
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.5567485882446512,
|
| 218 |
+
"grad_norm": 3.338602066040039,
|
| 219 |
+
"learning_rate": 0.00018157001114235688,
|
| 220 |
+
"loss": 0.3609,
|
| 221 |
+
"step": 14000
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.5766324663962459,
|
| 225 |
+
"grad_norm": 2.987213611602783,
|
| 226 |
+
"learning_rate": 0.0001809081020852125,
|
| 227 |
+
"loss": 0.3495,
|
| 228 |
+
"step": 14500
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 0.5965163445478406,
|
| 232 |
+
"grad_norm": 2.0543649196624756,
|
| 233 |
+
"learning_rate": 0.00018024619302806812,
|
| 234 |
+
"loss": 0.3504,
|
| 235 |
+
"step": 15000
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 0.6000954426151277,
|
| 239 |
+
"eval_loss": 0.33587488532066345,
|
| 240 |
+
"eval_runtime": 32.1839,
|
| 241 |
+
"eval_samples_per_second": 15.536,
|
| 242 |
+
"eval_steps_per_second": 7.768,
|
| 243 |
+
"step": 15090
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.6164002226994353,
|
| 247 |
+
"grad_norm": 2.5325777530670166,
|
| 248 |
+
"learning_rate": 0.00017958295749986736,
|
| 249 |
+
"loss": 0.3457,
|
| 250 |
+
"step": 15500
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.6362841008510299,
|
| 254 |
+
"grad_norm": 1.9482051134109497,
|
| 255 |
+
"learning_rate": 0.00017891972197166657,
|
| 256 |
+
"loss": 0.3556,
|
| 257 |
+
"step": 16000
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.6561679790026247,
|
| 261 |
+
"grad_norm": 2.1341960430145264,
|
| 262 |
+
"learning_rate": 0.0001782564864434658,
|
| 263 |
+
"loss": 0.3416,
|
| 264 |
+
"step": 16500
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.6760518571542193,
|
| 268 |
+
"grad_norm": 1.5148736238479614,
|
| 269 |
+
"learning_rate": 0.00017759325091526505,
|
| 270 |
+
"loss": 0.3419,
|
| 271 |
+
"step": 17000
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.6959357353058141,
|
| 275 |
+
"grad_norm": 3.0605080127716064,
|
| 276 |
+
"learning_rate": 0.00017693001538706426,
|
| 277 |
+
"loss": 0.3362,
|
| 278 |
+
"step": 17500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.7158196134574087,
|
| 282 |
+
"grad_norm": 1.941981315612793,
|
| 283 |
+
"learning_rate": 0.00017626677985886348,
|
| 284 |
+
"loss": 0.3309,
|
| 285 |
+
"step": 18000
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.7357034916090034,
|
| 289 |
+
"grad_norm": 3.0862131118774414,
|
| 290 |
+
"learning_rate": 0.00017560354433066272,
|
| 291 |
+
"loss": 0.3388,
|
| 292 |
+
"step": 18500
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 0.7555873697605981,
|
| 296 |
+
"grad_norm": 2.0479063987731934,
|
| 297 |
+
"learning_rate": 0.00017494163527351835,
|
| 298 |
+
"loss": 0.3307,
|
| 299 |
+
"step": 19000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.7754712479121928,
|
| 303 |
+
"grad_norm": 1.3942956924438477,
|
| 304 |
+
"learning_rate": 0.00017427839974531756,
|
| 305 |
+
"loss": 0.3333,
|
| 306 |
+
"step": 19500
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 0.7953551260637874,
|
| 310 |
+
"grad_norm": 1.5307426452636719,
|
| 311 |
+
"learning_rate": 0.00017361516421711678,
|
| 312 |
+
"loss": 0.3352,
|
| 313 |
+
"step": 20000
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.8001272568201702,
|
| 317 |
+
"eval_loss": 0.312864750623703,
|
| 318 |
+
"eval_runtime": 32.2758,
|
| 319 |
+
"eval_samples_per_second": 15.491,
|
| 320 |
+
"eval_steps_per_second": 7.746,
|
| 321 |
+
"step": 20120
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.8152390042153822,
|
| 325 |
+
"grad_norm": 2.5831427574157715,
|
| 326 |
+
"learning_rate": 0.00017295192868891602,
|
| 327 |
+
"loss": 0.3377,
|
| 328 |
+
"step": 20500
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.8351228823669768,
|
| 332 |
+
"grad_norm": 1.8084157705307007,
|
| 333 |
+
"learning_rate": 0.00017229001963177165,
|
| 334 |
+
"loss": 0.3227,
|
| 335 |
+
"step": 21000
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 0.8550067605185715,
|
| 339 |
+
"grad_norm": 0.7025062441825867,
|
| 340 |
+
"learning_rate": 0.00017162678410357086,
|
| 341 |
+
"loss": 0.3166,
|
| 342 |
+
"step": 21500
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 0.8748906386701663,
|
| 346 |
+
"grad_norm": 3.2739861011505127,
|
| 347 |
+
"learning_rate": 0.0001709635485753701,
|
| 348 |
+
"loss": 0.3127,
|
| 349 |
+
"step": 22000
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 0.8947745168217609,
|
| 353 |
+
"grad_norm": 1.438970685005188,
|
| 354 |
+
"learning_rate": 0.00017030031304716932,
|
| 355 |
+
"loss": 0.3238,
|
| 356 |
+
"step": 22500
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.9146583949733555,
|
| 360 |
+
"grad_norm": 1.9412046670913696,
|
| 361 |
+
"learning_rate": 0.00016963840399002495,
|
| 362 |
+
"loss": 0.3266,
|
| 363 |
+
"step": 23000
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 0.9345422731249503,
|
| 367 |
+
"grad_norm": 0.4293102025985718,
|
| 368 |
+
"learning_rate": 0.00016897516846182416,
|
| 369 |
+
"loss": 0.3324,
|
| 370 |
+
"step": 23500
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.954426151276545,
|
| 374 |
+
"grad_norm": 2.2440125942230225,
|
| 375 |
+
"learning_rate": 0.0001683119329336234,
|
| 376 |
+
"loss": 0.3206,
|
| 377 |
+
"step": 24000
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.9743100294281397,
|
| 381 |
+
"grad_norm": 2.241762399673462,
|
| 382 |
+
"learning_rate": 0.00016764869740542261,
|
| 383 |
+
"loss": 0.3078,
|
| 384 |
+
"step": 24500
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.9941939075797344,
|
| 388 |
+
"grad_norm": 2.4710514545440674,
|
| 389 |
+
"learning_rate": 0.00016698546187722183,
|
| 390 |
+
"loss": 0.3186,
|
| 391 |
+
"step": 25000
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 1.0001590710252128,
|
| 395 |
+
"eval_loss": 0.30107319355010986,
|
| 396 |
+
"eval_runtime": 32.3666,
|
| 397 |
+
"eval_samples_per_second": 15.448,
|
| 398 |
+
"eval_steps_per_second": 7.724,
|
| 399 |
+
"step": 25150
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 1.014077785731329,
|
| 403 |
+
"grad_norm": 0.9671458601951599,
|
| 404 |
+
"learning_rate": 0.0001663222263490211,
|
| 405 |
+
"loss": 0.277,
|
| 406 |
+
"step": 25500
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 1.0339616638829237,
|
| 410 |
+
"grad_norm": 1.7514474391937256,
|
| 411 |
+
"learning_rate": 0.0001656589908208203,
|
| 412 |
+
"loss": 0.2747,
|
| 413 |
+
"step": 26000
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 1.0538455420345185,
|
| 417 |
+
"grad_norm": 2.764012336730957,
|
| 418 |
+
"learning_rate": 0.00016499575529261952,
|
| 419 |
+
"loss": 0.2803,
|
| 420 |
+
"step": 26500
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 1.0737294201861132,
|
| 424 |
+
"grad_norm": 2.6387948989868164,
|
| 425 |
+
"learning_rate": 0.00016433384623547515,
|
| 426 |
+
"loss": 0.2757,
|
| 427 |
+
"step": 27000
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 1.0936132983377078,
|
| 431 |
+
"grad_norm": 3.577725410461426,
|
| 432 |
+
"learning_rate": 0.0001636706107072744,
|
| 433 |
+
"loss": 0.2819,
|
| 434 |
+
"step": 27500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 1.1134971764893025,
|
| 438 |
+
"grad_norm": 2.584761619567871,
|
| 439 |
+
"learning_rate": 0.0001630073751790736,
|
| 440 |
+
"loss": 0.2874,
|
| 441 |
+
"step": 28000
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 1.1333810546408971,
|
| 445 |
+
"grad_norm": 3.0081098079681396,
|
| 446 |
+
"learning_rate": 0.00016234413965087282,
|
| 447 |
+
"loss": 0.2804,
|
| 448 |
+
"step": 28500
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 1.1532649327924918,
|
| 452 |
+
"grad_norm": 2.1544599533081055,
|
| 453 |
+
"learning_rate": 0.00016168223059372845,
|
| 454 |
+
"loss": 0.2801,
|
| 455 |
+
"step": 29000
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 1.1731488109440866,
|
| 459 |
+
"grad_norm": 5.006481170654297,
|
| 460 |
+
"learning_rate": 0.00016101899506552767,
|
| 461 |
+
"loss": 0.2844,
|
| 462 |
+
"step": 29500
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 1.1930326890956813,
|
| 466 |
+
"grad_norm": 1.6833091974258423,
|
| 467 |
+
"learning_rate": 0.0001603557595373269,
|
| 468 |
+
"loss": 0.2763,
|
| 469 |
+
"step": 30000
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 1.2001908852302554,
|
| 473 |
+
"eval_loss": 0.2905046343803406,
|
| 474 |
+
"eval_runtime": 31.2019,
|
| 475 |
+
"eval_samples_per_second": 16.025,
|
| 476 |
+
"eval_steps_per_second": 8.012,
|
| 477 |
+
"step": 30180
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 1.212916567247276,
|
| 481 |
+
"grad_norm": 2.382586717605591,
|
| 482 |
+
"learning_rate": 0.00015969252400912614,
|
| 483 |
+
"loss": 0.2677,
|
| 484 |
+
"step": 30500
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 1.2328004453988706,
|
| 488 |
+
"grad_norm": 1.8768630027770996,
|
| 489 |
+
"learning_rate": 0.00015902928848092536,
|
| 490 |
+
"loss": 0.2695,
|
| 491 |
+
"step": 31000
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 1.2526843235504652,
|
| 495 |
+
"grad_norm": 2.0774290561676025,
|
| 496 |
+
"learning_rate": 0.00015836605295272457,
|
| 497 |
+
"loss": 0.2775,
|
| 498 |
+
"step": 31500
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 1.27256820170206,
|
| 502 |
+
"grad_norm": 1.6370826959609985,
|
| 503 |
+
"learning_rate": 0.0001577028174245238,
|
| 504 |
+
"loss": 0.2734,
|
| 505 |
+
"step": 32000
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 1.2924520798536547,
|
| 509 |
+
"grad_norm": 2.671633720397949,
|
| 510 |
+
"learning_rate": 0.00015704090836737944,
|
| 511 |
+
"loss": 0.2787,
|
| 512 |
+
"step": 32500
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 1.3123359580052494,
|
| 516 |
+
"grad_norm": 2.7384798526763916,
|
| 517 |
+
"learning_rate": 0.00015637767283917866,
|
| 518 |
+
"loss": 0.281,
|
| 519 |
+
"step": 33000
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 1.332219836156844,
|
| 523 |
+
"grad_norm": 2.3242549896240234,
|
| 524 |
+
"learning_rate": 0.00015571443731097787,
|
| 525 |
+
"loss": 0.2731,
|
| 526 |
+
"step": 33500
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 1.3521037143084387,
|
| 530 |
+
"grad_norm": 2.6705803871154785,
|
| 531 |
+
"learning_rate": 0.0001550512017827771,
|
| 532 |
+
"loss": 0.2743,
|
| 533 |
+
"step": 34000
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 1.3719875924600333,
|
| 537 |
+
"grad_norm": 2.173910617828369,
|
| 538 |
+
"learning_rate": 0.00015438796625457632,
|
| 539 |
+
"loss": 0.2719,
|
| 540 |
+
"step": 34500
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 1.391871470611628,
|
| 544 |
+
"grad_norm": 1.6040276288986206,
|
| 545 |
+
"learning_rate": 0.00015372473072637556,
|
| 546 |
+
"loss": 0.2711,
|
| 547 |
+
"step": 35000
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 1.400222699435298,
|
| 551 |
+
"eval_loss": 0.27763494849205017,
|
| 552 |
+
"eval_runtime": 32.7842,
|
| 553 |
+
"eval_samples_per_second": 15.251,
|
| 554 |
+
"eval_steps_per_second": 7.626,
|
| 555 |
+
"step": 35210
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.4117553487632228,
|
| 559 |
+
"grad_norm": 1.9321585893630981,
|
| 560 |
+
"learning_rate": 0.00015306149519817477,
|
| 561 |
+
"loss": 0.2779,
|
| 562 |
+
"step": 35500
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.4316392269148175,
|
| 566 |
+
"grad_norm": 2.574788808822632,
|
| 567 |
+
"learning_rate": 0.000152398259669974,
|
| 568 |
+
"loss": 0.2699,
|
| 569 |
+
"step": 36000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.4515231050664121,
|
| 573 |
+
"grad_norm": 1.1774181127548218,
|
| 574 |
+
"learning_rate": 0.00015173635061282962,
|
| 575 |
+
"loss": 0.2649,
|
| 576 |
+
"step": 36500
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.4714069832180068,
|
| 580 |
+
"grad_norm": 1.9039517641067505,
|
| 581 |
+
"learning_rate": 0.00015107311508462886,
|
| 582 |
+
"loss": 0.2672,
|
| 583 |
+
"step": 37000
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.4912908613696017,
|
| 587 |
+
"grad_norm": 2.026780605316162,
|
| 588 |
+
"learning_rate": 0.0001504098795564281,
|
| 589 |
+
"loss": 0.2644,
|
| 590 |
+
"step": 37500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.5111747395211963,
|
| 594 |
+
"grad_norm": 1.4502995014190674,
|
| 595 |
+
"learning_rate": 0.0001497466440282273,
|
| 596 |
+
"loss": 0.2715,
|
| 597 |
+
"step": 38000
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.531058617672791,
|
| 601 |
+
"grad_norm": 1.9970542192459106,
|
| 602 |
+
"learning_rate": 0.00014908340850002653,
|
| 603 |
+
"loss": 0.2674,
|
| 604 |
+
"step": 38500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.5509424958243856,
|
| 608 |
+
"grad_norm": 0.804521918296814,
|
| 609 |
+
"learning_rate": 0.00014842149944288216,
|
| 610 |
+
"loss": 0.2631,
|
| 611 |
+
"step": 39000
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.5708263739759802,
|
| 615 |
+
"grad_norm": 3.0539793968200684,
|
| 616 |
+
"learning_rate": 0.0001477582639146814,
|
| 617 |
+
"loss": 0.2629,
|
| 618 |
+
"step": 39500
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.5907102521275749,
|
| 622 |
+
"grad_norm": 2.970489025115967,
|
| 623 |
+
"learning_rate": 0.0001470950283864806,
|
| 624 |
+
"loss": 0.2579,
|
| 625 |
+
"step": 40000
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.6002545136403405,
|
| 629 |
+
"eval_loss": 0.27427446842193604,
|
| 630 |
+
"eval_runtime": 31.1724,
|
| 631 |
+
"eval_samples_per_second": 16.04,
|
| 632 |
+
"eval_steps_per_second": 8.02,
|
| 633 |
+
"step": 40240
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 1.6105941302791695,
|
| 637 |
+
"grad_norm": 2.624372959136963,
|
| 638 |
+
"learning_rate": 0.00014643179285827982,
|
| 639 |
+
"loss": 0.2533,
|
| 640 |
+
"step": 40500
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 1.6304780084307642,
|
| 644 |
+
"grad_norm": 1.7556920051574707,
|
| 645 |
+
"learning_rate": 0.00014576988380113549,
|
| 646 |
+
"loss": 0.2652,
|
| 647 |
+
"step": 41000
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 1.650361886582359,
|
| 651 |
+
"grad_norm": 2.7402307987213135,
|
| 652 |
+
"learning_rate": 0.0001451066482729347,
|
| 653 |
+
"loss": 0.2579,
|
| 654 |
+
"step": 41500
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 1.6702457647339537,
|
| 658 |
+
"grad_norm": 2.745375871658325,
|
| 659 |
+
"learning_rate": 0.0001444434127447339,
|
| 660 |
+
"loss": 0.2591,
|
| 661 |
+
"step": 42000
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 1.6901296428855483,
|
| 665 |
+
"grad_norm": 1.3609963655471802,
|
| 666 |
+
"learning_rate": 0.00014378017721653315,
|
| 667 |
+
"loss": 0.2557,
|
| 668 |
+
"step": 42500
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 1.7100135210371432,
|
| 672 |
+
"grad_norm": 3.024712324142456,
|
| 673 |
+
"learning_rate": 0.00014311826815938879,
|
| 674 |
+
"loss": 0.269,
|
| 675 |
+
"step": 43000
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 1.7298973991887379,
|
| 679 |
+
"grad_norm": 3.284465789794922,
|
| 680 |
+
"learning_rate": 0.000142455032631188,
|
| 681 |
+
"loss": 0.266,
|
| 682 |
+
"step": 43500
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 1.7497812773403325,
|
| 686 |
+
"grad_norm": 1.9791799783706665,
|
| 687 |
+
"learning_rate": 0.0001417917971029872,
|
| 688 |
+
"loss": 0.2718,
|
| 689 |
+
"step": 44000
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 1.7696651554919272,
|
| 693 |
+
"grad_norm": 1.614481806755066,
|
| 694 |
+
"learning_rate": 0.00014112856157478645,
|
| 695 |
+
"loss": 0.2579,
|
| 696 |
+
"step": 44500
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 1.7895490336435218,
|
| 700 |
+
"grad_norm": 2.6444413661956787,
|
| 701 |
+
"learning_rate": 0.00014046665251764206,
|
| 702 |
+
"loss": 0.2484,
|
| 703 |
+
"step": 45000
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 1.8002863278453831,
|
| 707 |
+
"eval_loss": 0.2599533200263977,
|
| 708 |
+
"eval_runtime": 30.2149,
|
| 709 |
+
"eval_samples_per_second": 16.548,
|
| 710 |
+
"eval_steps_per_second": 8.274,
|
| 711 |
+
"step": 45270
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 1.8094329117951164,
|
| 715 |
+
"grad_norm": 1.9396257400512695,
|
| 716 |
+
"learning_rate": 0.0001398034169894413,
|
| 717 |
+
"loss": 0.2512,
|
| 718 |
+
"step": 45500
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 1.829316789946711,
|
| 722 |
+
"grad_norm": 1.0529255867004395,
|
| 723 |
+
"learning_rate": 0.00013914018146124054,
|
| 724 |
+
"loss": 0.25,
|
| 725 |
+
"step": 46000
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 1.8492006680983057,
|
| 729 |
+
"grad_norm": 1.858811378479004,
|
| 730 |
+
"learning_rate": 0.00013847694593303975,
|
| 731 |
+
"loss": 0.2689,
|
| 732 |
+
"step": 46500
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 1.8690845462499006,
|
| 736 |
+
"grad_norm": 1.4326385259628296,
|
| 737 |
+
"learning_rate": 0.00013781371040483896,
|
| 738 |
+
"loss": 0.2539,
|
| 739 |
+
"step": 47000
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 1.8889684244014953,
|
| 743 |
+
"grad_norm": 4.373912811279297,
|
| 744 |
+
"learning_rate": 0.0001371518013476946,
|
| 745 |
+
"loss": 0.2614,
|
| 746 |
+
"step": 47500
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 1.90885230255309,
|
| 750 |
+
"grad_norm": 1.9298418760299683,
|
| 751 |
+
"learning_rate": 0.00013648856581949384,
|
| 752 |
+
"loss": 0.2539,
|
| 753 |
+
"step": 48000
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 1.9287361807046848,
|
| 757 |
+
"grad_norm": 1.697535753250122,
|
| 758 |
+
"learning_rate": 0.00013582665676234944,
|
| 759 |
+
"loss": 0.2606,
|
| 760 |
+
"step": 48500
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 1.9486200588562794,
|
| 764 |
+
"grad_norm": 2.6713733673095703,
|
| 765 |
+
"learning_rate": 0.00013516342123414866,
|
| 766 |
+
"loss": 0.2538,
|
| 767 |
+
"step": 49000
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 1.968503937007874,
|
| 771 |
+
"grad_norm": 4.244944095611572,
|
| 772 |
+
"learning_rate": 0.00013450018570594792,
|
| 773 |
+
"loss": 0.256,
|
| 774 |
+
"step": 49500
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 1.9883878151594687,
|
| 778 |
+
"grad_norm": 2.484851360321045,
|
| 779 |
+
"learning_rate": 0.00013383695017774714,
|
| 780 |
+
"loss": 0.263,
|
| 781 |
+
"step": 50000
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 2.0003181420504257,
|
| 785 |
+
"eval_loss": 0.26018011569976807,
|
| 786 |
+
"eval_runtime": 32.8317,
|
| 787 |
+
"eval_samples_per_second": 15.229,
|
| 788 |
+
"eval_steps_per_second": 7.615,
|
| 789 |
+
"step": 50300
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 2.0082716933110634,
|
| 793 |
+
"grad_norm": 2.1330833435058594,
|
| 794 |
+
"learning_rate": 0.00013317371464954635,
|
| 795 |
+
"loss": 0.2315,
|
| 796 |
+
"step": 50500
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 2.028155571462658,
|
| 800 |
+
"grad_norm": 2.0739586353302,
|
| 801 |
+
"learning_rate": 0.0001325104791213456,
|
| 802 |
+
"loss": 0.2178,
|
| 803 |
+
"step": 51000
|
| 804 |
+
},
|
| 805 |
+
{
|
| 806 |
+
"epoch": 2.0480394496142527,
|
| 807 |
+
"grad_norm": 2.4237098693847656,
|
| 808 |
+
"learning_rate": 0.0001318472435931448,
|
| 809 |
+
"loss": 0.2178,
|
| 810 |
+
"step": 51500
|
| 811 |
+
},
|
| 812 |
+
{
|
| 813 |
+
"epoch": 2.0679233277658473,
|
| 814 |
+
"grad_norm": 1.9835237264633179,
|
| 815 |
+
"learning_rate": 0.00013118533453600044,
|
| 816 |
+
"loss": 0.2102,
|
| 817 |
+
"step": 52000
|
| 818 |
+
},
|
| 819 |
+
{
|
| 820 |
+
"epoch": 2.087807205917442,
|
| 821 |
+
"grad_norm": 4.415806770324707,
|
| 822 |
+
"learning_rate": 0.00013052209900779965,
|
| 823 |
+
"loss": 0.2194,
|
| 824 |
+
"step": 52500
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 2.107691084069037,
|
| 828 |
+
"grad_norm": 2.1202263832092285,
|
| 829 |
+
"learning_rate": 0.0001298588634795989,
|
| 830 |
+
"loss": 0.2172,
|
| 831 |
+
"step": 53000
|
| 832 |
+
},
|
| 833 |
+
{
|
| 834 |
+
"epoch": 2.1275749622206317,
|
| 835 |
+
"grad_norm": 1.4174340963363647,
|
| 836 |
+
"learning_rate": 0.0001291956279513981,
|
| 837 |
+
"loss": 0.2256,
|
| 838 |
+
"step": 53500
|
| 839 |
+
},
|
| 840 |
+
{
|
| 841 |
+
"epoch": 2.1474588403722263,
|
| 842 |
+
"grad_norm": 2.1882033348083496,
|
| 843 |
+
"learning_rate": 0.0001285323924231973,
|
| 844 |
+
"loss": 0.2219,
|
| 845 |
+
"step": 54000
|
| 846 |
+
},
|
| 847 |
+
{
|
| 848 |
+
"epoch": 2.167342718523821,
|
| 849 |
+
"grad_norm": 1.7076680660247803,
|
| 850 |
+
"learning_rate": 0.00012787048336605297,
|
| 851 |
+
"loss": 0.2222,
|
| 852 |
+
"step": 54500
|
| 853 |
+
},
|
| 854 |
+
{
|
| 855 |
+
"epoch": 2.1872265966754156,
|
| 856 |
+
"grad_norm": 3.283176898956299,
|
| 857 |
+
"learning_rate": 0.0001272072478378522,
|
| 858 |
+
"loss": 0.2108,
|
| 859 |
+
"step": 55000
|
| 860 |
+
},
|
| 861 |
+
{
|
| 862 |
+
"epoch": 2.2003499562554683,
|
| 863 |
+
"eval_loss": 0.25801968574523926,
|
| 864 |
+
"eval_runtime": 31.5914,
|
| 865 |
+
"eval_samples_per_second": 15.827,
|
| 866 |
+
"eval_steps_per_second": 7.914,
|
| 867 |
+
"step": 55330
|
| 868 |
+
},
|
| 869 |
+
{
|
| 870 |
+
"epoch": 2.2071104748270103,
|
| 871 |
+
"grad_norm": 1.5287063121795654,
|
| 872 |
+
"learning_rate": 0.0001265440123096514,
|
| 873 |
+
"loss": 0.2237,
|
| 874 |
+
"step": 55500
|
| 875 |
+
},
|
| 876 |
+
{
|
| 877 |
+
"epoch": 2.226994352978605,
|
| 878 |
+
"grad_norm": 3.3277740478515625,
|
| 879 |
+
"learning_rate": 0.00012588077678145064,
|
| 880 |
+
"loss": 0.2202,
|
| 881 |
+
"step": 56000
|
| 882 |
+
},
|
| 883 |
+
{
|
| 884 |
+
"epoch": 2.2468782311301996,
|
| 885 |
+
"grad_norm": 2.4081342220306396,
|
| 886 |
+
"learning_rate": 0.00012521754125324988,
|
| 887 |
+
"loss": 0.225,
|
| 888 |
+
"step": 56500
|
| 889 |
+
},
|
| 890 |
+
{
|
| 891 |
+
"epoch": 2.2667621092817942,
|
| 892 |
+
"grad_norm": 5.641383171081543,
|
| 893 |
+
"learning_rate": 0.0001245543057250491,
|
| 894 |
+
"loss": 0.229,
|
| 895 |
+
"step": 57000
|
| 896 |
+
},
|
| 897 |
+
{
|
| 898 |
+
"epoch": 2.286645987433389,
|
| 899 |
+
"grad_norm": 1.27823007106781,
|
| 900 |
+
"learning_rate": 0.0001238910701968483,
|
| 901 |
+
"loss": 0.224,
|
| 902 |
+
"step": 57500
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 2.3065298655849835,
|
| 906 |
+
"grad_norm": 1.7878299951553345,
|
| 907 |
+
"learning_rate": 0.00012322783466864754,
|
| 908 |
+
"loss": 0.221,
|
| 909 |
+
"step": 58000
|
| 910 |
+
},
|
| 911 |
+
{
|
| 912 |
+
"epoch": 2.3264137437365786,
|
| 913 |
+
"grad_norm": 4.454659938812256,
|
| 914 |
+
"learning_rate": 0.00012256725208255957,
|
| 915 |
+
"loss": 0.2226,
|
| 916 |
+
"step": 58500
|
| 917 |
+
},
|
| 918 |
+
{
|
| 919 |
+
"epoch": 2.3462976218881733,
|
| 920 |
+
"grad_norm": 6.880167007446289,
|
| 921 |
+
"learning_rate": 0.00012190401655435879,
|
| 922 |
+
"loss": 0.2233,
|
| 923 |
+
"step": 59000
|
| 924 |
+
},
|
| 925 |
+
{
|
| 926 |
+
"epoch": 2.366181500039768,
|
| 927 |
+
"grad_norm": 3.2891809940338135,
|
| 928 |
+
"learning_rate": 0.00012124078102615801,
|
| 929 |
+
"loss": 0.2253,
|
| 930 |
+
"step": 59500
|
| 931 |
+
},
|
| 932 |
+
{
|
| 933 |
+
"epoch": 2.3860653781913626,
|
| 934 |
+
"grad_norm": 1.4356240034103394,
|
| 935 |
+
"learning_rate": 0.00012057754549795724,
|
| 936 |
+
"loss": 0.2026,
|
| 937 |
+
"step": 60000
|
| 938 |
+
},
|
| 939 |
+
{
|
| 940 |
+
"epoch": 2.400381770460511,
|
| 941 |
+
"eval_loss": 0.2525613307952881,
|
| 942 |
+
"eval_runtime": 31.802,
|
| 943 |
+
"eval_samples_per_second": 15.722,
|
| 944 |
+
"eval_steps_per_second": 7.861,
|
| 945 |
+
"step": 60360
|
| 946 |
+
},
|
| 947 |
+
{
|
| 948 |
+
"epoch": 2.405949256342957,
|
| 949 |
+
"grad_norm": 3.3116261959075928,
|
| 950 |
+
"learning_rate": 0.00011991430996975645,
|
| 951 |
+
"loss": 0.2137,
|
| 952 |
+
"step": 60500
|
| 953 |
+
},
|
| 954 |
+
{
|
| 955 |
+
"epoch": 2.425833134494552,
|
| 956 |
+
"grad_norm": 5.441344738006592,
|
| 957 |
+
"learning_rate": 0.0001192524009126121,
|
| 958 |
+
"loss": 0.2111,
|
| 959 |
+
"step": 61000
|
| 960 |
+
},
|
| 961 |
+
{
|
| 962 |
+
"epoch": 2.4457170126461465,
|
| 963 |
+
"grad_norm": 3.4003379344940186,
|
| 964 |
+
"learning_rate": 0.00011858916538441131,
|
| 965 |
+
"loss": 0.2074,
|
| 966 |
+
"step": 61500
|
| 967 |
+
},
|
| 968 |
+
{
|
| 969 |
+
"epoch": 2.465600890797741,
|
| 970 |
+
"grad_norm": 3.139822483062744,
|
| 971 |
+
"learning_rate": 0.00011792592985621054,
|
| 972 |
+
"loss": 0.2292,
|
| 973 |
+
"step": 62000
|
| 974 |
+
},
|
| 975 |
+
{
|
| 976 |
+
"epoch": 2.485484768949336,
|
| 977 |
+
"grad_norm": 2.3518755435943604,
|
| 978 |
+
"learning_rate": 0.00011726269432800976,
|
| 979 |
+
"loss": 0.2179,
|
| 980 |
+
"step": 62500
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 2.5053686471009304,
|
| 984 |
+
"grad_norm": 1.8685603141784668,
|
| 985 |
+
"learning_rate": 0.000116599458799809,
|
| 986 |
+
"loss": 0.2142,
|
| 987 |
+
"step": 63000
|
| 988 |
+
},
|
| 989 |
+
{
|
| 990 |
+
"epoch": 2.525252525252525,
|
| 991 |
+
"grad_norm": 1.3411154747009277,
|
| 992 |
+
"learning_rate": 0.00011593622327160823,
|
| 993 |
+
"loss": 0.213,
|
| 994 |
+
"step": 63500
|
| 995 |
+
},
|
| 996 |
+
{
|
| 997 |
+
"epoch": 2.54513640340412,
|
| 998 |
+
"grad_norm": 2.8563320636749268,
|
| 999 |
+
"learning_rate": 0.00011527298774340744,
|
| 1000 |
+
"loss": 0.2143,
|
| 1001 |
+
"step": 64000
|
| 1002 |
+
},
|
| 1003 |
+
{
|
| 1004 |
+
"epoch": 2.5650202815557144,
|
| 1005 |
+
"grad_norm": 1.2390650510787964,
|
| 1006 |
+
"learning_rate": 0.00011460975221520667,
|
| 1007 |
+
"loss": 0.2043,
|
| 1008 |
+
"step": 64500
|
| 1009 |
+
},
|
| 1010 |
+
{
|
| 1011 |
+
"epoch": 2.5849041597073095,
|
| 1012 |
+
"grad_norm": 5.118180274963379,
|
| 1013 |
+
"learning_rate": 0.0001139491696291187,
|
| 1014 |
+
"loss": 0.2228,
|
| 1015 |
+
"step": 65000
|
| 1016 |
+
},
|
| 1017 |
+
{
|
| 1018 |
+
"epoch": 2.600413584665553,
|
| 1019 |
+
"eval_loss": 0.2472601681947708,
|
| 1020 |
+
"eval_runtime": 30.4328,
|
| 1021 |
+
"eval_samples_per_second": 16.43,
|
| 1022 |
+
"eval_steps_per_second": 8.215,
|
| 1023 |
+
"step": 65390
|
| 1024 |
+
},
|
| 1025 |
+
{
|
| 1026 |
+
"epoch": 2.604788037858904,
|
| 1027 |
+
"grad_norm": 1.907989501953125,
|
| 1028 |
+
"learning_rate": 0.00011328593410091792,
|
| 1029 |
+
"loss": 0.2227,
|
| 1030 |
+
"step": 65500
|
| 1031 |
+
},
|
| 1032 |
+
{
|
| 1033 |
+
"epoch": 2.6246719160104988,
|
| 1034 |
+
"grad_norm": 1.7330621480941772,
|
| 1035 |
+
"learning_rate": 0.00011262269857271715,
|
| 1036 |
+
"loss": 0.2246,
|
| 1037 |
+
"step": 66000
|
| 1038 |
+
},
|
| 1039 |
+
{
|
| 1040 |
+
"epoch": 2.6445557941620934,
|
| 1041 |
+
"grad_norm": 2.17523193359375,
|
| 1042 |
+
"learning_rate": 0.00011196078951557278,
|
| 1043 |
+
"loss": 0.2198,
|
| 1044 |
+
"step": 66500
|
| 1045 |
+
},
|
| 1046 |
+
{
|
| 1047 |
+
"epoch": 2.664439672313688,
|
| 1048 |
+
"grad_norm": 3.0923678874969482,
|
| 1049 |
+
"learning_rate": 0.00011129755398737201,
|
| 1050 |
+
"loss": 0.2117,
|
| 1051 |
+
"step": 67000
|
| 1052 |
+
},
|
| 1053 |
+
{
|
| 1054 |
+
"epoch": 2.6843235504652827,
|
| 1055 |
+
"grad_norm": 1.7811450958251953,
|
| 1056 |
+
"learning_rate": 0.00011063431845917122,
|
| 1057 |
+
"loss": 0.2281,
|
| 1058 |
+
"step": 67500
|
| 1059 |
+
},
|
| 1060 |
+
{
|
| 1061 |
+
"epoch": 2.7042074286168774,
|
| 1062 |
+
"grad_norm": 3.345266580581665,
|
| 1063 |
+
"learning_rate": 0.00010997108293097045,
|
| 1064 |
+
"loss": 0.2175,
|
| 1065 |
+
"step": 68000
|
| 1066 |
+
},
|
| 1067 |
+
{
|
| 1068 |
+
"epoch": 2.724091306768472,
|
| 1069 |
+
"grad_norm": 2.302898406982422,
|
| 1070 |
+
"learning_rate": 0.00010930784740276967,
|
| 1071 |
+
"loss": 0.221,
|
| 1072 |
+
"step": 68500
|
| 1073 |
+
},
|
| 1074 |
+
{
|
| 1075 |
+
"epoch": 2.7439751849200666,
|
| 1076 |
+
"grad_norm": 1.6563972234725952,
|
| 1077 |
+
"learning_rate": 0.00010864461187456889,
|
| 1078 |
+
"loss": 0.211,
|
| 1079 |
+
"step": 69000
|
| 1080 |
+
},
|
| 1081 |
+
{
|
| 1082 |
+
"epoch": 2.7638590630716617,
|
| 1083 |
+
"grad_norm": 1.6703941822052002,
|
| 1084 |
+
"learning_rate": 0.00010798137634636814,
|
| 1085 |
+
"loss": 0.2089,
|
| 1086 |
+
"step": 69500
|
| 1087 |
+
},
|
| 1088 |
+
{
|
| 1089 |
+
"epoch": 2.783742941223256,
|
| 1090 |
+
"grad_norm": 2.7034361362457275,
|
| 1091 |
+
"learning_rate": 0.00010731814081816735,
|
| 1092 |
+
"loss": 0.2187,
|
| 1093 |
+
"step": 70000
|
| 1094 |
+
},
|
| 1095 |
+
{
|
| 1096 |
+
"epoch": 2.800445398870596,
|
| 1097 |
+
"eval_loss": 0.23865090310573578,
|
| 1098 |
+
"eval_runtime": 32.3036,
|
| 1099 |
+
"eval_samples_per_second": 15.478,
|
| 1100 |
+
"eval_steps_per_second": 7.739,
|
| 1101 |
+
"step": 70420
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 2.803626819374851,
|
| 1105 |
+
"grad_norm": 3.077442169189453,
|
| 1106 |
+
"learning_rate": 0.00010665623176102297,
|
| 1107 |
+
"loss": 0.213,
|
| 1108 |
+
"step": 70500
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 2.8235106975264457,
|
| 1112 |
+
"grad_norm": 2.5723166465759277,
|
| 1113 |
+
"learning_rate": 0.0001059929962328222,
|
| 1114 |
+
"loss": 0.2147,
|
| 1115 |
+
"step": 71000
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 2.8433945756780403,
|
| 1119 |
+
"grad_norm": 3.936866521835327,
|
| 1120 |
+
"learning_rate": 0.00010532976070462144,
|
| 1121 |
+
"loss": 0.2121,
|
| 1122 |
+
"step": 71500
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 2.863278453829635,
|
| 1126 |
+
"grad_norm": 1.5824639797210693,
|
| 1127 |
+
"learning_rate": 0.00010466652517642067,
|
| 1128 |
+
"loss": 0.2151,
|
| 1129 |
+
"step": 72000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 2.8831623319812296,
|
| 1133 |
+
"grad_norm": 0.3409406244754791,
|
| 1134 |
+
"learning_rate": 0.00010400328964821988,
|
| 1135 |
+
"loss": 0.211,
|
| 1136 |
+
"step": 72500
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 2.9030462101328243,
|
| 1140 |
+
"grad_norm": 2.319749355316162,
|
| 1141 |
+
"learning_rate": 0.0001033400541200191,
|
| 1142 |
+
"loss": 0.211,
|
| 1143 |
+
"step": 73000
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 2.922930088284419,
|
| 1147 |
+
"grad_norm": 4.800658226013184,
|
| 1148 |
+
"learning_rate": 0.00010267681859181833,
|
| 1149 |
+
"loss": 0.2075,
|
| 1150 |
+
"step": 73500
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 2.9428139664360136,
|
| 1154 |
+
"grad_norm": 2.598496675491333,
|
| 1155 |
+
"learning_rate": 0.00010201358306361754,
|
| 1156 |
+
"loss": 0.2128,
|
| 1157 |
+
"step": 74000
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 2.962697844587608,
|
| 1161 |
+
"grad_norm": 4.022119998931885,
|
| 1162 |
+
"learning_rate": 0.0001013503475354168,
|
| 1163 |
+
"loss": 0.2184,
|
| 1164 |
+
"step": 74500
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 2.9825817227392033,
|
| 1168 |
+
"grad_norm": 3.838801145553589,
|
| 1169 |
+
"learning_rate": 0.0001006897649493288,
|
| 1170 |
+
"loss": 0.2053,
|
| 1171 |
+
"step": 75000
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 3.000477213075638,
|
| 1175 |
+
"eval_loss": 0.2344769537448883,
|
| 1176 |
+
"eval_runtime": 31.4754,
|
| 1177 |
+
"eval_samples_per_second": 15.885,
|
| 1178 |
+
"eval_steps_per_second": 7.943,
|
| 1179 |
+
"step": 75450
|
| 1180 |
+
},
|
| 1181 |
+
{
|
| 1182 |
+
"epoch": 3.002465600890798,
|
| 1183 |
+
"grad_norm": 2.6975882053375244,
|
| 1184 |
+
"learning_rate": 0.00010002652942112805,
|
| 1185 |
+
"loss": 0.1945,
|
| 1186 |
+
"step": 75500
|
| 1187 |
+
},
|
| 1188 |
+
{
|
| 1189 |
+
"epoch": 3.0223494790423926,
|
| 1190 |
+
"grad_norm": 3.4830374717712402,
|
| 1191 |
+
"learning_rate": 9.936329389292725e-05,
|
| 1192 |
+
"loss": 0.1727,
|
| 1193 |
+
"step": 76000
|
| 1194 |
+
},
|
| 1195 |
+
{
|
| 1196 |
+
"epoch": 3.0422333571939872,
|
| 1197 |
+
"grad_norm": 2.481358289718628,
|
| 1198 |
+
"learning_rate": 9.870005836472649e-05,
|
| 1199 |
+
"loss": 0.1748,
|
| 1200 |
+
"step": 76500
|
| 1201 |
+
},
|
| 1202 |
+
{
|
| 1203 |
+
"epoch": 3.062117235345582,
|
| 1204 |
+
"grad_norm": 0.8453766107559204,
|
| 1205 |
+
"learning_rate": 9.803682283652572e-05,
|
| 1206 |
+
"loss": 0.1801,
|
| 1207 |
+
"step": 77000
|
| 1208 |
+
},
|
| 1209 |
+
{
|
| 1210 |
+
"epoch": 3.0820011134971765,
|
| 1211 |
+
"grad_norm": 2.1420843601226807,
|
| 1212 |
+
"learning_rate": 9.737358730832493e-05,
|
| 1213 |
+
"loss": 0.173,
|
| 1214 |
+
"step": 77500
|
| 1215 |
+
},
|
| 1216 |
+
{
|
| 1217 |
+
"epoch": 3.101884991648771,
|
| 1218 |
+
"grad_norm": 2.596755266189575,
|
| 1219 |
+
"learning_rate": 9.671035178012417e-05,
|
| 1220 |
+
"loss": 0.1852,
|
| 1221 |
+
"step": 78000
|
| 1222 |
+
},
|
| 1223 |
+
{
|
| 1224 |
+
"epoch": 3.121768869800366,
|
| 1225 |
+
"grad_norm": 1.8939439058303833,
|
| 1226 |
+
"learning_rate": 9.604711625192338e-05,
|
| 1227 |
+
"loss": 0.1859,
|
| 1228 |
+
"step": 78500
|
| 1229 |
+
},
|
| 1230 |
+
{
|
| 1231 |
+
"epoch": 3.1416527479519605,
|
| 1232 |
+
"grad_norm": 2.0201351642608643,
|
| 1233 |
+
"learning_rate": 9.538520719477902e-05,
|
| 1234 |
+
"loss": 0.183,
|
| 1235 |
+
"step": 79000
|
| 1236 |
+
},
|
| 1237 |
+
{
|
| 1238 |
+
"epoch": 3.161536626103555,
|
| 1239 |
+
"grad_norm": 2.634225368499756,
|
| 1240 |
+
"learning_rate": 9.472197166657823e-05,
|
| 1241 |
+
"loss": 0.1756,
|
| 1242 |
+
"step": 79500
|
| 1243 |
+
},
|
| 1244 |
+
{
|
| 1245 |
+
"epoch": 3.1814205042551498,
|
| 1246 |
+
"grad_norm": 0.12749190628528595,
|
| 1247 |
+
"learning_rate": 9.405873613837747e-05,
|
| 1248 |
+
"loss": 0.1761,
|
| 1249 |
+
"step": 80000
|
| 1250 |
+
},
|
| 1251 |
+
{
|
| 1252 |
+
"epoch": 3.2005090272806807,
|
| 1253 |
+
"eval_loss": 0.23787811398506165,
|
| 1254 |
+
"eval_runtime": 30.3497,
|
| 1255 |
+
"eval_samples_per_second": 16.475,
|
| 1256 |
+
"eval_steps_per_second": 8.237,
|
| 1257 |
+
"step": 80480
|
| 1258 |
+
},
|
| 1259 |
+
{
|
| 1260 |
+
"epoch": 3.2013043824067444,
|
| 1261 |
+
"grad_norm": 2.9797298908233643,
|
| 1262 |
+
"learning_rate": 9.33955006101767e-05,
|
| 1263 |
+
"loss": 0.1795,
|
| 1264 |
+
"step": 80500
|
| 1265 |
+
},
|
| 1266 |
+
{
|
| 1267 |
+
"epoch": 3.221188260558339,
|
| 1268 |
+
"grad_norm": 2.532785177230835,
|
| 1269 |
+
"learning_rate": 9.273359155303232e-05,
|
| 1270 |
+
"loss": 0.1856,
|
| 1271 |
+
"step": 81000
|
| 1272 |
+
},
|
| 1273 |
+
{
|
| 1274 |
+
"epoch": 3.241072138709934,
|
| 1275 |
+
"grad_norm": 3.591059446334839,
|
| 1276 |
+
"learning_rate": 9.207035602483154e-05,
|
| 1277 |
+
"loss": 0.178,
|
| 1278 |
+
"step": 81500
|
| 1279 |
+
},
|
| 1280 |
+
{
|
| 1281 |
+
"epoch": 3.260956016861529,
|
| 1282 |
+
"grad_norm": 1.8293282985687256,
|
| 1283 |
+
"learning_rate": 9.140712049663077e-05,
|
| 1284 |
+
"loss": 0.1845,
|
| 1285 |
+
"step": 82000
|
| 1286 |
+
},
|
| 1287 |
+
{
|
| 1288 |
+
"epoch": 3.2808398950131235,
|
| 1289 |
+
"grad_norm": 3.8247385025024414,
|
| 1290 |
+
"learning_rate": 9.07452114394864e-05,
|
| 1291 |
+
"loss": 0.1921,
|
| 1292 |
+
"step": 82500
|
| 1293 |
+
},
|
| 1294 |
+
{
|
| 1295 |
+
"epoch": 3.300723773164718,
|
| 1296 |
+
"grad_norm": 3.090575695037842,
|
| 1297 |
+
"learning_rate": 9.008197591128561e-05,
|
| 1298 |
+
"loss": 0.1725,
|
| 1299 |
+
"step": 83000
|
| 1300 |
+
},
|
| 1301 |
+
{
|
| 1302 |
+
"epoch": 3.3206076513163127,
|
| 1303 |
+
"grad_norm": 2.8974783420562744,
|
| 1304 |
+
"learning_rate": 8.941874038308484e-05,
|
| 1305 |
+
"loss": 0.1838,
|
| 1306 |
+
"step": 83500
|
| 1307 |
+
},
|
| 1308 |
+
{
|
| 1309 |
+
"epoch": 3.3404915294679074,
|
| 1310 |
+
"grad_norm": 4.1438469886779785,
|
| 1311 |
+
"learning_rate": 8.875550485488408e-05,
|
| 1312 |
+
"loss": 0.1832,
|
| 1313 |
+
"step": 84000
|
| 1314 |
+
},
|
| 1315 |
+
{
|
| 1316 |
+
"epoch": 3.360375407619502,
|
| 1317 |
+
"grad_norm": 2.026402235031128,
|
| 1318 |
+
"learning_rate": 8.809226932668329e-05,
|
| 1319 |
+
"loss": 0.1787,
|
| 1320 |
+
"step": 84500
|
| 1321 |
+
},
|
| 1322 |
+
{
|
| 1323 |
+
"epoch": 3.3802592857710967,
|
| 1324 |
+
"grad_norm": 2.8283655643463135,
|
| 1325 |
+
"learning_rate": 8.742903379848252e-05,
|
| 1326 |
+
"loss": 0.1833,
|
| 1327 |
+
"step": 85000
|
| 1328 |
+
},
|
| 1329 |
+
{
|
| 1330 |
+
"epoch": 3.4001431639226913,
|
| 1331 |
+
"grad_norm": 3.4132397174835205,
|
| 1332 |
+
"learning_rate": 8.676712474133814e-05,
|
| 1333 |
+
"loss": 0.1773,
|
| 1334 |
+
"step": 85500
|
| 1335 |
+
},
|
| 1336 |
+
{
|
| 1337 |
+
"epoch": 3.400540841485723,
|
| 1338 |
+
"eval_loss": 0.23229274153709412,
|
| 1339 |
+
"eval_runtime": 30.3693,
|
| 1340 |
+
"eval_samples_per_second": 16.464,
|
| 1341 |
+
"eval_steps_per_second": 8.232,
|
| 1342 |
+
"step": 85510
|
| 1343 |
+
},
|
| 1344 |
+
{
|
| 1345 |
+
"epoch": 3.420027042074286,
|
| 1346 |
+
"grad_norm": 1.775317668914795,
|
| 1347 |
+
"learning_rate": 8.610388921313738e-05,
|
| 1348 |
+
"loss": 0.1794,
|
| 1349 |
+
"step": 86000
|
| 1350 |
+
},
|
| 1351 |
+
{
|
| 1352 |
+
"epoch": 3.4399109202258806,
|
| 1353 |
+
"grad_norm": 3.800232172012329,
|
| 1354 |
+
"learning_rate": 8.54406536849366e-05,
|
| 1355 |
+
"loss": 0.1812,
|
| 1356 |
+
"step": 86500
|
| 1357 |
+
},
|
| 1358 |
+
{
|
| 1359 |
+
"epoch": 3.4597947983774757,
|
| 1360 |
+
"grad_norm": 2.0861783027648926,
|
| 1361 |
+
"learning_rate": 8.477741815673582e-05,
|
| 1362 |
+
"loss": 0.1788,
|
| 1363 |
+
"step": 87000
|
| 1364 |
+
},
|
| 1365 |
+
{
|
| 1366 |
+
"epoch": 3.4796786765290704,
|
| 1367 |
+
"grad_norm": 2.620603561401367,
|
| 1368 |
+
"learning_rate": 8.411418262853506e-05,
|
| 1369 |
+
"loss": 0.1711,
|
| 1370 |
+
"step": 87500
|
| 1371 |
+
},
|
| 1372 |
+
{
|
| 1373 |
+
"epoch": 3.499562554680665,
|
| 1374 |
+
"grad_norm": 3.5984723567962646,
|
| 1375 |
+
"learning_rate": 8.345094710033427e-05,
|
| 1376 |
+
"loss": 0.1819,
|
| 1377 |
+
"step": 88000
|
| 1378 |
+
},
|
| 1379 |
+
{
|
| 1380 |
+
"epoch": 3.5194464328322597,
|
| 1381 |
+
"grad_norm": 2.1630051136016846,
|
| 1382 |
+
"learning_rate": 8.27877115721335e-05,
|
| 1383 |
+
"loss": 0.179,
|
| 1384 |
+
"step": 88500
|
| 1385 |
+
},
|
| 1386 |
+
{
|
| 1387 |
+
"epoch": 3.5393303109838543,
|
| 1388 |
+
"grad_norm": 2.205453395843506,
|
| 1389 |
+
"learning_rate": 8.212447604393274e-05,
|
| 1390 |
+
"loss": 0.1758,
|
| 1391 |
+
"step": 89000
|
| 1392 |
+
},
|
| 1393 |
+
{
|
| 1394 |
+
"epoch": 3.559214189135449,
|
| 1395 |
+
"grad_norm": 1.9129152297973633,
|
| 1396 |
+
"learning_rate": 8.146256698678834e-05,
|
| 1397 |
+
"loss": 0.1784,
|
| 1398 |
+
"step": 89500
|
| 1399 |
+
},
|
| 1400 |
+
{
|
| 1401 |
+
"epoch": 3.5790980672870436,
|
| 1402 |
+
"grad_norm": 1.8072373867034912,
|
| 1403 |
+
"learning_rate": 8.079933145858758e-05,
|
| 1404 |
+
"loss": 0.1795,
|
| 1405 |
+
"step": 90000
|
| 1406 |
+
},
|
| 1407 |
+
{
|
| 1408 |
+
"epoch": 3.5989819454386383,
|
| 1409 |
+
"grad_norm": 1.9712532758712769,
|
| 1410 |
+
"learning_rate": 8.01360959303868e-05,
|
| 1411 |
+
"loss": 0.1689,
|
| 1412 |
+
"step": 90500
|
| 1413 |
+
},
|
| 1414 |
+
{
|
| 1415 |
+
"epoch": 3.600572655690766,
|
| 1416 |
+
"eval_loss": 0.2264697253704071,
|
| 1417 |
+
"eval_runtime": 30.1359,
|
| 1418 |
+
"eval_samples_per_second": 16.591,
|
| 1419 |
+
"eval_steps_per_second": 8.296,
|
| 1420 |
+
"step": 90540
|
| 1421 |
+
},
|
| 1422 |
+
{
|
| 1423 |
+
"epoch": 3.618865823590233,
|
| 1424 |
+
"grad_norm": 3.026998996734619,
|
| 1425 |
+
"learning_rate": 7.947286040218604e-05,
|
| 1426 |
+
"loss": 0.1687,
|
| 1427 |
+
"step": 91000
|
| 1428 |
+
},
|
| 1429 |
+
{
|
| 1430 |
+
"epoch": 3.638749701741828,
|
| 1431 |
+
"grad_norm": 1.1603697538375854,
|
| 1432 |
+
"learning_rate": 7.880962487398525e-05,
|
| 1433 |
+
"loss": 0.1842,
|
| 1434 |
+
"step": 91500
|
| 1435 |
+
},
|
| 1436 |
+
{
|
| 1437 |
+
"epoch": 3.658633579893422,
|
| 1438 |
+
"grad_norm": 3.8162033557891846,
|
| 1439 |
+
"learning_rate": 7.814771581684088e-05,
|
| 1440 |
+
"loss": 0.1778,
|
| 1441 |
+
"step": 92000
|
| 1442 |
+
},
|
| 1443 |
+
{
|
| 1444 |
+
"epoch": 3.6785174580450173,
|
| 1445 |
+
"grad_norm": 2.1724276542663574,
|
| 1446 |
+
"learning_rate": 7.748448028864011e-05,
|
| 1447 |
+
"loss": 0.176,
|
| 1448 |
+
"step": 92500
|
| 1449 |
+
},
|
| 1450 |
+
{
|
| 1451 |
+
"epoch": 3.698401336196612,
|
| 1452 |
+
"grad_norm": 2.8712663650512695,
|
| 1453 |
+
"learning_rate": 7.682257123149573e-05,
|
| 1454 |
+
"loss": 0.1687,
|
| 1455 |
+
"step": 93000
|
| 1456 |
+
},
|
| 1457 |
+
{
|
| 1458 |
+
"epoch": 3.7182852143482066,
|
| 1459 |
+
"grad_norm": 3.8236796855926514,
|
| 1460 |
+
"learning_rate": 7.615933570329496e-05,
|
| 1461 |
+
"loss": 0.1772,
|
| 1462 |
+
"step": 93500
|
| 1463 |
+
},
|
| 1464 |
+
{
|
| 1465 |
+
"epoch": 3.7381690924998012,
|
| 1466 |
+
"grad_norm": 2.69323468208313,
|
| 1467 |
+
"learning_rate": 7.549610017509418e-05,
|
| 1468 |
+
"loss": 0.171,
|
| 1469 |
+
"step": 94000
|
| 1470 |
+
},
|
| 1471 |
+
{
|
| 1472 |
+
"epoch": 3.758052970651396,
|
| 1473 |
+
"grad_norm": 4.058226585388184,
|
| 1474 |
+
"learning_rate": 7.483286464689341e-05,
|
| 1475 |
+
"loss": 0.1713,
|
| 1476 |
+
"step": 94500
|
| 1477 |
+
},
|
| 1478 |
+
{
|
| 1479 |
+
"epoch": 3.7779368488029905,
|
| 1480 |
+
"grad_norm": 1.3760497570037842,
|
| 1481 |
+
"learning_rate": 7.416962911869263e-05,
|
| 1482 |
+
"loss": 0.1739,
|
| 1483 |
+
"step": 95000
|
| 1484 |
+
},
|
| 1485 |
+
{
|
| 1486 |
+
"epoch": 3.797820726954585,
|
| 1487 |
+
"grad_norm": 3.206932544708252,
|
| 1488 |
+
"learning_rate": 7.350639359049186e-05,
|
| 1489 |
+
"loss": 0.1707,
|
| 1490 |
+
"step": 95500
|
| 1491 |
+
},
|
| 1492 |
+
{
|
| 1493 |
+
"epoch": 3.8006044698958084,
|
| 1494 |
+
"eval_loss": 0.22563844919204712,
|
| 1495 |
+
"eval_runtime": 30.9046,
|
| 1496 |
+
"eval_samples_per_second": 16.179,
|
| 1497 |
+
"eval_steps_per_second": 8.089,
|
| 1498 |
+
"step": 95570
|
| 1499 |
+
},
|
| 1500 |
+
{
|
| 1501 |
+
"epoch": 3.81770460510618,
|
| 1502 |
+
"grad_norm": 1.1313693523406982,
|
| 1503 |
+
"learning_rate": 7.28444845333475e-05,
|
| 1504 |
+
"loss": 0.1804,
|
| 1505 |
+
"step": 96000
|
| 1506 |
+
},
|
| 1507 |
+
{
|
| 1508 |
+
"epoch": 3.8375884832577745,
|
| 1509 |
+
"grad_norm": 6.103767395019531,
|
| 1510 |
+
"learning_rate": 7.218124900514671e-05,
|
| 1511 |
+
"loss": 0.1838,
|
| 1512 |
+
"step": 96500
|
| 1513 |
+
},
|
| 1514 |
+
{
|
| 1515 |
+
"epoch": 3.8574723614093696,
|
| 1516 |
+
"grad_norm": 2.6996240615844727,
|
| 1517 |
+
"learning_rate": 7.151801347694593e-05,
|
| 1518 |
+
"loss": 0.1775,
|
| 1519 |
+
"step": 97000
|
| 1520 |
+
},
|
| 1521 |
+
{
|
| 1522 |
+
"epoch": 3.8773562395609638,
|
| 1523 |
+
"grad_norm": 3.337559700012207,
|
| 1524 |
+
"learning_rate": 7.085477794874516e-05,
|
| 1525 |
+
"loss": 0.1746,
|
| 1526 |
+
"step": 97500
|
| 1527 |
+
},
|
| 1528 |
+
{
|
| 1529 |
+
"epoch": 3.897240117712559,
|
| 1530 |
+
"grad_norm": 2.2691235542297363,
|
| 1531 |
+
"learning_rate": 7.019154242054439e-05,
|
| 1532 |
+
"loss": 0.1699,
|
| 1533 |
+
"step": 98000
|
| 1534 |
+
},
|
| 1535 |
+
{
|
| 1536 |
+
"epoch": 3.9171239958641535,
|
| 1537 |
+
"grad_norm": 1.5143208503723145,
|
| 1538 |
+
"learning_rate": 6.952830689234361e-05,
|
| 1539 |
+
"loss": 0.1733,
|
| 1540 |
+
"step": 98500
|
| 1541 |
+
},
|
| 1542 |
+
{
|
| 1543 |
+
"epoch": 3.937007874015748,
|
| 1544 |
+
"grad_norm": 1.51828932762146,
|
| 1545 |
+
"learning_rate": 6.886507136414284e-05,
|
| 1546 |
+
"loss": 0.1703,
|
| 1547 |
+
"step": 99000
|
| 1548 |
+
},
|
| 1549 |
+
{
|
| 1550 |
+
"epoch": 3.956891752167343,
|
| 1551 |
+
"grad_norm": 2.325024127960205,
|
| 1552 |
+
"learning_rate": 6.820183583594206e-05,
|
| 1553 |
+
"loss": 0.1717,
|
| 1554 |
+
"step": 99500
|
| 1555 |
+
},
|
| 1556 |
+
{
|
| 1557 |
+
"epoch": 3.9767756303189374,
|
| 1558 |
+
"grad_norm": 1.8743467330932617,
|
| 1559 |
+
"learning_rate": 6.753860030774129e-05,
|
| 1560 |
+
"loss": 0.1711,
|
| 1561 |
+
"step": 100000
|
| 1562 |
+
},
|
| 1563 |
+
{
|
| 1564 |
+
"epoch": 3.996659508470532,
|
| 1565 |
+
"grad_norm": 6.472159385681152,
|
| 1566 |
+
"learning_rate": 6.687669125059691e-05,
|
| 1567 |
+
"loss": 0.1764,
|
| 1568 |
+
"step": 100500
|
| 1569 |
+
},
|
| 1570 |
+
{
|
| 1571 |
+
"epoch": 4.000636284100851,
|
| 1572 |
+
"eval_loss": 0.21969762444496155,
|
| 1573 |
+
"eval_runtime": 30.6282,
|
| 1574 |
+
"eval_samples_per_second": 16.325,
|
| 1575 |
+
"eval_steps_per_second": 8.162,
|
| 1576 |
+
"step": 100600
|
| 1577 |
+
},
|
| 1578 |
+
{
|
| 1579 |
+
"epoch": 4.016543386622127,
|
| 1580 |
+
"grad_norm": 0.35814252495765686,
|
| 1581 |
+
"learning_rate": 6.621345572239615e-05,
|
| 1582 |
+
"loss": 0.1495,
|
| 1583 |
+
"step": 101000
|
| 1584 |
+
},
|
| 1585 |
+
{
|
| 1586 |
+
"epoch": 4.036427264773722,
|
| 1587 |
+
"grad_norm": 1.2356332540512085,
|
| 1588 |
+
"learning_rate": 6.555022019419536e-05,
|
| 1589 |
+
"loss": 0.1439,
|
| 1590 |
+
"step": 101500
|
| 1591 |
+
},
|
| 1592 |
+
{
|
| 1593 |
+
"epoch": 4.056311142925316,
|
| 1594 |
+
"grad_norm": 2.1953282356262207,
|
| 1595 |
+
"learning_rate": 6.488698466599459e-05,
|
| 1596 |
+
"loss": 0.1506,
|
| 1597 |
+
"step": 102000
|
| 1598 |
+
},
|
| 1599 |
+
{
|
| 1600 |
+
"epoch": 4.076195021076911,
|
| 1601 |
+
"grad_norm": 3.411731004714966,
|
| 1602 |
+
"learning_rate": 6.422640207990662e-05,
|
| 1603 |
+
"loss": 0.1526,
|
| 1604 |
+
"step": 102500
|
| 1605 |
+
},
|
| 1606 |
+
{
|
| 1607 |
+
"epoch": 4.096078899228505,
|
| 1608 |
+
"grad_norm": 1.3503237962722778,
|
| 1609 |
+
"learning_rate": 6.356316655170585e-05,
|
| 1610 |
+
"loss": 0.1411,
|
| 1611 |
+
"step": 103000
|
| 1612 |
+
},
|
| 1613 |
+
{
|
| 1614 |
+
"epoch": 4.1159627773801,
|
| 1615 |
+
"grad_norm": 4.093628406524658,
|
| 1616 |
+
"learning_rate": 6.289993102350507e-05,
|
| 1617 |
+
"loss": 0.1434,
|
| 1618 |
+
"step": 103500
|
| 1619 |
+
},
|
| 1620 |
+
{
|
| 1621 |
+
"epoch": 4.135846655531695,
|
| 1622 |
+
"grad_norm": 2.112595319747925,
|
| 1623 |
+
"learning_rate": 6.22366954953043e-05,
|
| 1624 |
+
"loss": 0.1446,
|
| 1625 |
+
"step": 104000
|
| 1626 |
+
},
|
| 1627 |
+
{
|
| 1628 |
+
"epoch": 4.15573053368329,
|
| 1629 |
+
"grad_norm": 3.1277015209198,
|
| 1630 |
+
"learning_rate": 6.157345996710352e-05,
|
| 1631 |
+
"loss": 0.1471,
|
| 1632 |
+
"step": 104500
|
| 1633 |
+
},
|
| 1634 |
+
{
|
| 1635 |
+
"epoch": 4.175614411834884,
|
| 1636 |
+
"grad_norm": 2.3667871952056885,
|
| 1637 |
+
"learning_rate": 6.0911550909959145e-05,
|
| 1638 |
+
"loss": 0.1471,
|
| 1639 |
+
"step": 105000
|
| 1640 |
+
},
|
| 1641 |
+
{
|
| 1642 |
+
"epoch": 4.195498289986479,
|
| 1643 |
+
"grad_norm": 2.765684127807617,
|
| 1644 |
+
"learning_rate": 6.024831538175837e-05,
|
| 1645 |
+
"loss": 0.1377,
|
| 1646 |
+
"step": 105500
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 4.2006680983058935,
|
| 1650 |
+
"eval_loss": 0.22619232535362244,
|
| 1651 |
+
"eval_runtime": 30.6831,
|
| 1652 |
+
"eval_samples_per_second": 16.296,
|
| 1653 |
+
"eval_steps_per_second": 8.148,
|
| 1654 |
+
"step": 105630
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 4.215382168138074,
|
| 1658 |
+
"grad_norm": 2.138153314590454,
|
| 1659 |
+
"learning_rate": 5.9585079853557604e-05,
|
| 1660 |
+
"loss": 0.1384,
|
| 1661 |
+
"step": 106000
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 4.235266046289668,
|
| 1665 |
+
"grad_norm": 3.0117478370666504,
|
| 1666 |
+
"learning_rate": 5.892184432535682e-05,
|
| 1667 |
+
"loss": 0.1464,
|
| 1668 |
+
"step": 106500
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 4.255149924441263,
|
| 1672 |
+
"grad_norm": 1.6394137144088745,
|
| 1673 |
+
"learning_rate": 5.825860879715604e-05,
|
| 1674 |
+
"loss": 0.1524,
|
| 1675 |
+
"step": 107000
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 4.275033802592858,
|
| 1679 |
+
"grad_norm": 2.5397379398345947,
|
| 1680 |
+
"learning_rate": 5.759669974001167e-05,
|
| 1681 |
+
"loss": 0.1412,
|
| 1682 |
+
"step": 107500
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 4.294917680744453,
|
| 1686 |
+
"grad_norm": 3.301124095916748,
|
| 1687 |
+
"learning_rate": 5.69334642118109e-05,
|
| 1688 |
+
"loss": 0.1481,
|
| 1689 |
+
"step": 108000
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 4.314801558896047,
|
| 1693 |
+
"grad_norm": 1.5573328733444214,
|
| 1694 |
+
"learning_rate": 5.627022868361013e-05,
|
| 1695 |
+
"loss": 0.1504,
|
| 1696 |
+
"step": 108500
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 4.334685437047642,
|
| 1700 |
+
"grad_norm": 2.6243669986724854,
|
| 1701 |
+
"learning_rate": 5.560699315540935e-05,
|
| 1702 |
+
"loss": 0.1499,
|
| 1703 |
+
"step": 109000
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 4.354569315199236,
|
| 1707 |
+
"grad_norm": 5.335040092468262,
|
| 1708 |
+
"learning_rate": 5.494375762720858e-05,
|
| 1709 |
+
"loss": 0.1467,
|
| 1710 |
+
"step": 109500
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 4.374453193350831,
|
| 1714 |
+
"grad_norm": 4.008390426635742,
|
| 1715 |
+
"learning_rate": 5.428184857006421e-05,
|
| 1716 |
+
"loss": 0.1467,
|
| 1717 |
+
"step": 110000
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 4.3943370715024255,
|
| 1721 |
+
"grad_norm": 1.7349259853363037,
|
| 1722 |
+
"learning_rate": 5.361861304186343e-05,
|
| 1723 |
+
"loss": 0.1485,
|
| 1724 |
+
"step": 110500
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 4.4006999125109365,
|
| 1728 |
+
"eval_loss": 0.22301064431667328,
|
| 1729 |
+
"eval_runtime": 32.1763,
|
| 1730 |
+
"eval_samples_per_second": 15.539,
|
| 1731 |
+
"eval_steps_per_second": 7.77,
|
| 1732 |
+
"step": 110660
|
| 1733 |
+
},
|
| 1734 |
+
{
|
| 1735 |
+
"epoch": 4.414220949654021,
|
| 1736 |
+
"grad_norm": 7.040079116821289,
|
| 1737 |
+
"learning_rate": 5.2955377513662655e-05,
|
| 1738 |
+
"loss": 0.1404,
|
| 1739 |
+
"step": 111000
|
| 1740 |
+
},
|
| 1741 |
+
{
|
| 1742 |
+
"epoch": 4.434104827805616,
|
| 1743 |
+
"grad_norm": 2.5272045135498047,
|
| 1744 |
+
"learning_rate": 5.229214198546189e-05,
|
| 1745 |
+
"loss": 0.1558,
|
| 1746 |
+
"step": 111500
|
| 1747 |
+
},
|
| 1748 |
+
{
|
| 1749 |
+
"epoch": 4.45398870595721,
|
| 1750 |
+
"grad_norm": 0.3252130448818207,
|
| 1751 |
+
"learning_rate": 5.162890645726111e-05,
|
| 1752 |
+
"loss": 0.1546,
|
| 1753 |
+
"step": 112000
|
| 1754 |
+
},
|
| 1755 |
+
{
|
| 1756 |
+
"epoch": 4.473872584108805,
|
| 1757 |
+
"grad_norm": 3.8579819202423096,
|
| 1758 |
+
"learning_rate": 5.0965670929060326e-05,
|
| 1759 |
+
"loss": 0.1526,
|
| 1760 |
+
"step": 112500
|
| 1761 |
+
},
|
| 1762 |
+
{
|
| 1763 |
+
"epoch": 4.493756462260399,
|
| 1764 |
+
"grad_norm": 2.077725648880005,
|
| 1765 |
+
"learning_rate": 5.0303761871915954e-05,
|
| 1766 |
+
"loss": 0.1353,
|
| 1767 |
+
"step": 113000
|
| 1768 |
+
},
|
| 1769 |
+
{
|
| 1770 |
+
"epoch": 4.513640340411994,
|
| 1771 |
+
"grad_norm": 3.343392848968506,
|
| 1772 |
+
"learning_rate": 4.964052634371518e-05,
|
| 1773 |
+
"loss": 0.129,
|
| 1774 |
+
"step": 113500
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 4.5335242185635884,
|
| 1778 |
+
"grad_norm": 1.1551505327224731,
|
| 1779 |
+
"learning_rate": 4.8977290815514406e-05,
|
| 1780 |
+
"loss": 0.1331,
|
| 1781 |
+
"step": 114000
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 4.5534080967151835,
|
| 1785 |
+
"grad_norm": 4.196703910827637,
|
| 1786 |
+
"learning_rate": 4.831405528731363e-05,
|
| 1787 |
+
"loss": 0.1422,
|
| 1788 |
+
"step": 114500
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 4.573291974866778,
|
| 1792 |
+
"grad_norm": 2.8865489959716797,
|
| 1793 |
+
"learning_rate": 4.765214623016926e-05,
|
| 1794 |
+
"loss": 0.1466,
|
| 1795 |
+
"step": 115000
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 4.593175853018373,
|
| 1799 |
+
"grad_norm": 1.843113660812378,
|
| 1800 |
+
"learning_rate": 4.6988910701968486e-05,
|
| 1801 |
+
"loss": 0.1464,
|
| 1802 |
+
"step": 115500
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 4.600731726715979,
|
| 1806 |
+
"eval_loss": 0.21611736714839935,
|
| 1807 |
+
"eval_runtime": 31.4994,
|
| 1808 |
+
"eval_samples_per_second": 15.873,
|
| 1809 |
+
"eval_steps_per_second": 7.937,
|
| 1810 |
+
"step": 115690
|
| 1811 |
+
},
|
| 1812 |
+
{
|
| 1813 |
+
"epoch": 4.613059731169967,
|
| 1814 |
+
"grad_norm": 2.9247336387634277,
|
| 1815 |
+
"learning_rate": 4.632567517376771e-05,
|
| 1816 |
+
"loss": 0.1386,
|
| 1817 |
+
"step": 116000
|
| 1818 |
+
},
|
| 1819 |
+
{
|
| 1820 |
+
"epoch": 4.632943609321562,
|
| 1821 |
+
"grad_norm": 1.6676517724990845,
|
| 1822 |
+
"learning_rate": 4.566243964556693e-05,
|
| 1823 |
+
"loss": 0.1397,
|
| 1824 |
+
"step": 116500
|
| 1825 |
+
},
|
| 1826 |
+
{
|
| 1827 |
+
"epoch": 4.652827487473157,
|
| 1828 |
+
"grad_norm": 4.6178131103515625,
|
| 1829 |
+
"learning_rate": 4.4999204117366165e-05,
|
| 1830 |
+
"loss": 0.1472,
|
| 1831 |
+
"step": 117000
|
| 1832 |
+
},
|
| 1833 |
+
{
|
| 1834 |
+
"epoch": 4.672711365624751,
|
| 1835 |
+
"grad_norm": 2.4779419898986816,
|
| 1836 |
+
"learning_rate": 4.433729506022179e-05,
|
| 1837 |
+
"loss": 0.1437,
|
| 1838 |
+
"step": 117500
|
| 1839 |
+
},
|
| 1840 |
+
{
|
| 1841 |
+
"epoch": 4.6925952437763465,
|
| 1842 |
+
"grad_norm": 0.12668299674987793,
|
| 1843 |
+
"learning_rate": 4.367405953202101e-05,
|
| 1844 |
+
"loss": 0.1376,
|
| 1845 |
+
"step": 118000
|
| 1846 |
+
},
|
| 1847 |
+
{
|
| 1848 |
+
"epoch": 4.712479121927941,
|
| 1849 |
+
"grad_norm": 2.201352596282959,
|
| 1850 |
+
"learning_rate": 4.301082400382024e-05,
|
| 1851 |
+
"loss": 0.1322,
|
| 1852 |
+
"step": 118500
|
| 1853 |
+
},
|
| 1854 |
+
{
|
| 1855 |
+
"epoch": 4.732363000079536,
|
| 1856 |
+
"grad_norm": 1.9356328248977661,
|
| 1857 |
+
"learning_rate": 4.2347588475619464e-05,
|
| 1858 |
+
"loss": 0.1406,
|
| 1859 |
+
"step": 119000
|
| 1860 |
+
},
|
| 1861 |
+
{
|
| 1862 |
+
"epoch": 4.75224687823113,
|
| 1863 |
+
"grad_norm": 2.280499219894409,
|
| 1864 |
+
"learning_rate": 4.168567941847509e-05,
|
| 1865 |
+
"loss": 0.1387,
|
| 1866 |
+
"step": 119500
|
| 1867 |
+
},
|
| 1868 |
+
{
|
| 1869 |
+
"epoch": 4.772130756382725,
|
| 1870 |
+
"grad_norm": 3.5399367809295654,
|
| 1871 |
+
"learning_rate": 4.102244389027432e-05,
|
| 1872 |
+
"loss": 0.1375,
|
| 1873 |
+
"step": 120000
|
| 1874 |
+
},
|
| 1875 |
+
{
|
| 1876 |
+
"epoch": 4.792014634534319,
|
| 1877 |
+
"grad_norm": 2.2342448234558105,
|
| 1878 |
+
"learning_rate": 4.0359208362073544e-05,
|
| 1879 |
+
"loss": 0.1521,
|
| 1880 |
+
"step": 120500
|
| 1881 |
+
},
|
| 1882 |
+
{
|
| 1883 |
+
"epoch": 4.800763540921022,
|
| 1884 |
+
"eval_loss": 0.2177634984254837,
|
| 1885 |
+
"eval_runtime": 30.9261,
|
| 1886 |
+
"eval_samples_per_second": 16.168,
|
| 1887 |
+
"eval_steps_per_second": 8.084,
|
| 1888 |
+
"step": 120720
|
| 1889 |
+
},
|
| 1890 |
+
{
|
| 1891 |
+
"epoch": 4.811898512685914,
|
| 1892 |
+
"grad_norm": 1.7616039514541626,
|
| 1893 |
+
"learning_rate": 3.969597283387277e-05,
|
| 1894 |
+
"loss": 0.1429,
|
| 1895 |
+
"step": 121000
|
| 1896 |
+
},
|
| 1897 |
+
{
|
| 1898 |
+
"epoch": 4.831782390837509,
|
| 1899 |
+
"grad_norm": 4.162886142730713,
|
| 1900 |
+
"learning_rate": 3.903273730567199e-05,
|
| 1901 |
+
"loss": 0.1356,
|
| 1902 |
+
"step": 121500
|
| 1903 |
+
},
|
| 1904 |
+
{
|
| 1905 |
+
"epoch": 4.851666268989104,
|
| 1906 |
+
"grad_norm": 4.022665023803711,
|
| 1907 |
+
"learning_rate": 3.8369501777471215e-05,
|
| 1908 |
+
"loss": 0.1425,
|
| 1909 |
+
"step": 122000
|
| 1910 |
+
},
|
| 1911 |
+
{
|
| 1912 |
+
"epoch": 4.871550147140699,
|
| 1913 |
+
"grad_norm": 2.1347572803497314,
|
| 1914 |
+
"learning_rate": 3.770626624927044e-05,
|
| 1915 |
+
"loss": 0.1403,
|
| 1916 |
+
"step": 122500
|
| 1917 |
+
},
|
| 1918 |
+
{
|
| 1919 |
+
"epoch": 4.891434025292293,
|
| 1920 |
+
"grad_norm": 4.412343502044678,
|
| 1921 |
+
"learning_rate": 3.704303072106967e-05,
|
| 1922 |
+
"loss": 0.1423,
|
| 1923 |
+
"step": 123000
|
| 1924 |
+
},
|
| 1925 |
+
{
|
| 1926 |
+
"epoch": 4.911317903443888,
|
| 1927 |
+
"grad_norm": 3.702019214630127,
|
| 1928 |
+
"learning_rate": 3.6381121663925295e-05,
|
| 1929 |
+
"loss": 0.1415,
|
| 1930 |
+
"step": 123500
|
| 1931 |
+
},
|
| 1932 |
+
{
|
| 1933 |
+
"epoch": 4.931201781595482,
|
| 1934 |
+
"grad_norm": 7.128276348114014,
|
| 1935 |
+
"learning_rate": 3.571788613572452e-05,
|
| 1936 |
+
"loss": 0.1322,
|
| 1937 |
+
"step": 124000
|
| 1938 |
+
},
|
| 1939 |
+
{
|
| 1940 |
+
"epoch": 4.951085659747077,
|
| 1941 |
+
"grad_norm": 2.2007031440734863,
|
| 1942 |
+
"learning_rate": 3.505465060752375e-05,
|
| 1943 |
+
"loss": 0.1343,
|
| 1944 |
+
"step": 124500
|
| 1945 |
+
},
|
| 1946 |
+
{
|
| 1947 |
+
"epoch": 4.970969537898672,
|
| 1948 |
+
"grad_norm": 2.8083972930908203,
|
| 1949 |
+
"learning_rate": 3.439141507932297e-05,
|
| 1950 |
+
"loss": 0.1326,
|
| 1951 |
+
"step": 125000
|
| 1952 |
+
},
|
| 1953 |
+
{
|
| 1954 |
+
"epoch": 4.990853416050267,
|
| 1955 |
+
"grad_norm": 0.3411719501018524,
|
| 1956 |
+
"learning_rate": 3.37281795511222e-05,
|
| 1957 |
+
"loss": 0.1457,
|
| 1958 |
+
"step": 125500
|
| 1959 |
+
},
|
| 1960 |
+
{
|
| 1961 |
+
"epoch": 5.000795355126064,
|
| 1962 |
+
"eval_loss": 0.21443423628807068,
|
| 1963 |
+
"eval_runtime": 33.8081,
|
| 1964 |
+
"eval_samples_per_second": 14.789,
|
| 1965 |
+
"eval_steps_per_second": 7.395,
|
| 1966 |
+
"step": 125750
|
| 1967 |
+
},
|
| 1968 |
+
{
|
| 1969 |
+
"epoch": 5.010737294201861,
|
| 1970 |
+
"grad_norm": 3.1173503398895264,
|
| 1971 |
+
"learning_rate": 3.3064944022921426e-05,
|
| 1972 |
+
"loss": 0.129,
|
| 1973 |
+
"step": 126000
|
| 1974 |
+
},
|
| 1975 |
+
{
|
| 1976 |
+
"epoch": 5.030621172353456,
|
| 1977 |
+
"grad_norm": 2.295043468475342,
|
| 1978 |
+
"learning_rate": 3.2401708494720645e-05,
|
| 1979 |
+
"loss": 0.1088,
|
| 1980 |
+
"step": 126500
|
| 1981 |
+
},
|
| 1982 |
+
{
|
| 1983 |
+
"epoch": 5.05050505050505,
|
| 1984 |
+
"grad_norm": 2.1319775581359863,
|
| 1985 |
+
"learning_rate": 3.173979943757627e-05,
|
| 1986 |
+
"loss": 0.1116,
|
| 1987 |
+
"step": 127000
|
| 1988 |
+
},
|
| 1989 |
+
{
|
| 1990 |
+
"epoch": 5.070388928656645,
|
| 1991 |
+
"grad_norm": 3.116173028945923,
|
| 1992 |
+
"learning_rate": 3.10765639093755e-05,
|
| 1993 |
+
"loss": 0.1026,
|
| 1994 |
+
"step": 127500
|
| 1995 |
+
},
|
| 1996 |
+
{
|
| 1997 |
+
"epoch": 5.0902728068082395,
|
| 1998 |
+
"grad_norm": 4.195283889770508,
|
| 1999 |
+
"learning_rate": 3.0413328381174722e-05,
|
| 2000 |
+
"loss": 0.1171,
|
| 2001 |
+
"step": 128000
|
| 2002 |
+
},
|
| 2003 |
+
{
|
| 2004 |
+
"epoch": 5.1101566849598345,
|
| 2005 |
+
"grad_norm": 1.1767369508743286,
|
| 2006 |
+
"learning_rate": 2.9750092852973948e-05,
|
| 2007 |
+
"loss": 0.1163,
|
| 2008 |
+
"step": 128500
|
| 2009 |
+
},
|
| 2010 |
+
{
|
| 2011 |
+
"epoch": 5.13004056311143,
|
| 2012 |
+
"grad_norm": 3.730426073074341,
|
| 2013 |
+
"learning_rate": 2.9086857324773174e-05,
|
| 2014 |
+
"loss": 0.1114,
|
| 2015 |
+
"step": 129000
|
| 2016 |
+
},
|
| 2017 |
+
{
|
| 2018 |
+
"epoch": 5.149924441263024,
|
| 2019 |
+
"grad_norm": 3.132356643676758,
|
| 2020 |
+
"learning_rate": 2.8424948267628805e-05,
|
| 2021 |
+
"loss": 0.1165,
|
| 2022 |
+
"step": 129500
|
| 2023 |
+
},
|
| 2024 |
+
{
|
| 2025 |
+
"epoch": 5.169808319414619,
|
| 2026 |
+
"grad_norm": 6.5695672035217285,
|
| 2027 |
+
"learning_rate": 2.7761712739428025e-05,
|
| 2028 |
+
"loss": 0.1135,
|
| 2029 |
+
"step": 130000
|
| 2030 |
+
},
|
| 2031 |
+
{
|
| 2032 |
+
"epoch": 5.189692197566213,
|
| 2033 |
+
"grad_norm": 2.7988929748535156,
|
| 2034 |
+
"learning_rate": 2.709847721122725e-05,
|
| 2035 |
+
"loss": 0.1146,
|
| 2036 |
+
"step": 130500
|
| 2037 |
+
},
|
| 2038 |
+
{
|
| 2039 |
+
"epoch": 5.200827169331106,
|
| 2040 |
+
"eval_loss": 0.222051203250885,
|
| 2041 |
+
"eval_runtime": 31.5914,
|
| 2042 |
+
"eval_samples_per_second": 15.827,
|
| 2043 |
+
"eval_steps_per_second": 7.914,
|
| 2044 |
+
"step": 130780
|
| 2045 |
+
},
|
| 2046 |
+
{
|
| 2047 |
+
"epoch": 5.209576075717808,
|
| 2048 |
+
"grad_norm": 4.5756988525390625,
|
| 2049 |
+
"learning_rate": 2.643524168302648e-05,
|
| 2050 |
+
"loss": 0.111,
|
| 2051 |
+
"step": 131000
|
| 2052 |
+
},
|
| 2053 |
+
{
|
| 2054 |
+
"epoch": 5.229459953869402,
|
| 2055 |
+
"grad_norm": 1.7040164470672607,
|
| 2056 |
+
"learning_rate": 2.57720061548257e-05,
|
| 2057 |
+
"loss": 0.1141,
|
| 2058 |
+
"step": 131500
|
| 2059 |
+
},
|
| 2060 |
+
{
|
| 2061 |
+
"epoch": 5.2493438320209975,
|
| 2062 |
+
"grad_norm": 2.915128469467163,
|
| 2063 |
+
"learning_rate": 2.5110097097681327e-05,
|
| 2064 |
+
"loss": 0.115,
|
| 2065 |
+
"step": 132000
|
| 2066 |
+
},
|
| 2067 |
+
{
|
| 2068 |
+
"epoch": 5.269227710172592,
|
| 2069 |
+
"grad_norm": 3.992319345474243,
|
| 2070 |
+
"learning_rate": 2.4446861569480557e-05,
|
| 2071 |
+
"loss": 0.1106,
|
| 2072 |
+
"step": 132500
|
| 2073 |
+
},
|
| 2074 |
+
{
|
| 2075 |
+
"epoch": 5.289111588324187,
|
| 2076 |
+
"grad_norm": 2.6975860595703125,
|
| 2077 |
+
"learning_rate": 2.378362604127978e-05,
|
| 2078 |
+
"loss": 0.1194,
|
| 2079 |
+
"step": 133000
|
| 2080 |
+
},
|
| 2081 |
+
{
|
| 2082 |
+
"epoch": 5.308995466475781,
|
| 2083 |
+
"grad_norm": 5.065980434417725,
|
| 2084 |
+
"learning_rate": 2.3120390513079006e-05,
|
| 2085 |
+
"loss": 0.1098,
|
| 2086 |
+
"step": 133500
|
| 2087 |
+
},
|
| 2088 |
+
{
|
| 2089 |
+
"epoch": 5.328879344627376,
|
| 2090 |
+
"grad_norm": 1.7343071699142456,
|
| 2091 |
+
"learning_rate": 2.2457154984878232e-05,
|
| 2092 |
+
"loss": 0.1096,
|
| 2093 |
+
"step": 134000
|
| 2094 |
+
},
|
| 2095 |
+
{
|
| 2096 |
+
"epoch": 5.348763222778971,
|
| 2097 |
+
"grad_norm": 2.005739212036133,
|
| 2098 |
+
"learning_rate": 2.1793919456677455e-05,
|
| 2099 |
+
"loss": 0.1118,
|
| 2100 |
+
"step": 134500
|
| 2101 |
+
},
|
| 2102 |
+
{
|
| 2103 |
+
"epoch": 5.368647100930565,
|
| 2104 |
+
"grad_norm": 3.6282753944396973,
|
| 2105 |
+
"learning_rate": 2.1130683928476684e-05,
|
| 2106 |
+
"loss": 0.1139,
|
| 2107 |
+
"step": 135000
|
| 2108 |
+
},
|
| 2109 |
+
{
|
| 2110 |
+
"epoch": 5.3885309790821605,
|
| 2111 |
+
"grad_norm": 2.168957233428955,
|
| 2112 |
+
"learning_rate": 2.046877487133231e-05,
|
| 2113 |
+
"loss": 0.1187,
|
| 2114 |
+
"step": 135500
|
| 2115 |
+
},
|
| 2116 |
+
{
|
| 2117 |
+
"epoch": 5.400858983536149,
|
| 2118 |
+
"eval_loss": 0.220447838306427,
|
| 2119 |
+
"eval_runtime": 33.9447,
|
| 2120 |
+
"eval_samples_per_second": 14.73,
|
| 2121 |
+
"eval_steps_per_second": 7.365,
|
| 2122 |
+
"step": 135810
|
| 2123 |
+
},
|
| 2124 |
+
{
|
| 2125 |
+
"epoch": 5.408414857233755,
|
| 2126 |
+
"grad_norm": 1.5377305746078491,
|
| 2127 |
+
"learning_rate": 1.9805539343131535e-05,
|
| 2128 |
+
"loss": 0.1085,
|
| 2129 |
+
"step": 136000
|
| 2130 |
+
},
|
| 2131 |
+
{
|
| 2132 |
+
"epoch": 5.42829873538535,
|
| 2133 |
+
"grad_norm": 6.779255390167236,
|
| 2134 |
+
"learning_rate": 1.914230381493076e-05,
|
| 2135 |
+
"loss": 0.1166,
|
| 2136 |
+
"step": 136500
|
| 2137 |
+
},
|
| 2138 |
+
{
|
| 2139 |
+
"epoch": 5.448182613536944,
|
| 2140 |
+
"grad_norm": 1.6060508489608765,
|
| 2141 |
+
"learning_rate": 1.8479068286729983e-05,
|
| 2142 |
+
"loss": 0.1135,
|
| 2143 |
+
"step": 137000
|
| 2144 |
+
},
|
| 2145 |
+
{
|
| 2146 |
+
"epoch": 5.468066491688539,
|
| 2147 |
+
"grad_norm": 3.5860373973846436,
|
| 2148 |
+
"learning_rate": 1.781583275852921e-05,
|
| 2149 |
+
"loss": 0.1105,
|
| 2150 |
+
"step": 137500
|
| 2151 |
+
},
|
| 2152 |
+
{
|
| 2153 |
+
"epoch": 5.487950369840133,
|
| 2154 |
+
"grad_norm": 1.9102075099945068,
|
| 2155 |
+
"learning_rate": 1.7153923701384837e-05,
|
| 2156 |
+
"loss": 0.112,
|
| 2157 |
+
"step": 138000
|
| 2158 |
+
},
|
| 2159 |
+
{
|
| 2160 |
+
"epoch": 5.507834247991728,
|
| 2161 |
+
"grad_norm": 0.8978993892669678,
|
| 2162 |
+
"learning_rate": 1.6490688173184063e-05,
|
| 2163 |
+
"loss": 0.116,
|
| 2164 |
+
"step": 138500
|
| 2165 |
+
},
|
| 2166 |
+
{
|
| 2167 |
+
"epoch": 5.5277181261433235,
|
| 2168 |
+
"grad_norm": 1.6748642921447754,
|
| 2169 |
+
"learning_rate": 1.5827452644983286e-05,
|
| 2170 |
+
"loss": 0.1121,
|
| 2171 |
+
"step": 139000
|
| 2172 |
+
},
|
| 2173 |
+
{
|
| 2174 |
+
"epoch": 5.547602004294918,
|
| 2175 |
+
"grad_norm": 2.604186773300171,
|
| 2176 |
+
"learning_rate": 1.5164217116782512e-05,
|
| 2177 |
+
"loss": 0.1146,
|
| 2178 |
+
"step": 139500
|
| 2179 |
+
},
|
| 2180 |
+
{
|
| 2181 |
+
"epoch": 5.567485882446512,
|
| 2182 |
+
"grad_norm": 2.35908842086792,
|
| 2183 |
+
"learning_rate": 1.4500981588581738e-05,
|
| 2184 |
+
"loss": 0.1112,
|
| 2185 |
+
"step": 140000
|
| 2186 |
+
},
|
| 2187 |
+
{
|
| 2188 |
+
"epoch": 5.587369760598107,
|
| 2189 |
+
"grad_norm": 1.971954107284546,
|
| 2190 |
+
"learning_rate": 1.3837746060380963e-05,
|
| 2191 |
+
"loss": 0.1041,
|
| 2192 |
+
"step": 140500
|
| 2193 |
+
},
|
| 2194 |
+
{
|
| 2195 |
+
"epoch": 5.600890797741192,
|
| 2196 |
+
"eval_loss": 0.22010314464569092,
|
| 2197 |
+
"eval_runtime": 31.7301,
|
| 2198 |
+
"eval_samples_per_second": 15.758,
|
| 2199 |
+
"eval_steps_per_second": 7.879,
|
| 2200 |
+
"step": 140840
|
| 2201 |
+
},
|
| 2202 |
+
{
|
| 2203 |
+
"epoch": 5.607253638749702,
|
| 2204 |
+
"grad_norm": 0.5583412051200867,
|
| 2205 |
+
"learning_rate": 1.317583700323659e-05,
|
| 2206 |
+
"loss": 0.11,
|
| 2207 |
+
"step": 141000
|
| 2208 |
+
},
|
| 2209 |
+
{
|
| 2210 |
+
"epoch": 5.627137516901296,
|
| 2211 |
+
"grad_norm": 2.2566328048706055,
|
| 2212 |
+
"learning_rate": 1.2512601475035815e-05,
|
| 2213 |
+
"loss": 0.1148,
|
| 2214 |
+
"step": 141500
|
| 2215 |
+
},
|
| 2216 |
+
{
|
| 2217 |
+
"epoch": 5.647021395052891,
|
| 2218 |
+
"grad_norm": 2.328498601913452,
|
| 2219 |
+
"learning_rate": 1.1849365946835041e-05,
|
| 2220 |
+
"loss": 0.1079,
|
| 2221 |
+
"step": 142000
|
| 2222 |
+
},
|
| 2223 |
+
{
|
| 2224 |
+
"epoch": 5.666905273204486,
|
| 2225 |
+
"grad_norm": 1.9184355735778809,
|
| 2226 |
+
"learning_rate": 1.1186130418634267e-05,
|
| 2227 |
+
"loss": 0.115,
|
| 2228 |
+
"step": 142500
|
| 2229 |
+
},
|
| 2230 |
+
{
|
| 2231 |
+
"epoch": 5.686789151356081,
|
| 2232 |
+
"grad_norm": 2.273148536682129,
|
| 2233 |
+
"learning_rate": 1.052289489043349e-05,
|
| 2234 |
+
"loss": 0.1087,
|
| 2235 |
+
"step": 143000
|
| 2236 |
+
},
|
| 2237 |
+
{
|
| 2238 |
+
"epoch": 5.706673029507675,
|
| 2239 |
+
"grad_norm": 0.8123636245727539,
|
| 2240 |
+
"learning_rate": 9.859659362232716e-06,
|
| 2241 |
+
"loss": 0.1157,
|
| 2242 |
+
"step": 143500
|
| 2243 |
+
},
|
| 2244 |
+
{
|
| 2245 |
+
"epoch": 5.72655690765927,
|
| 2246 |
+
"grad_norm": 1.5079761743545532,
|
| 2247 |
+
"learning_rate": 9.196423834031942e-06,
|
| 2248 |
+
"loss": 0.1121,
|
| 2249 |
+
"step": 144000
|
| 2250 |
+
},
|
| 2251 |
+
{
|
| 2252 |
+
"epoch": 5.746440785810865,
|
| 2253 |
+
"grad_norm": 3.0341477394104004,
|
| 2254 |
+
"learning_rate": 8.533188305831168e-06,
|
| 2255 |
+
"loss": 0.1173,
|
| 2256 |
+
"step": 144500
|
| 2257 |
+
},
|
| 2258 |
+
{
|
| 2259 |
+
"epoch": 5.766324663962459,
|
| 2260 |
+
"grad_norm": 4.0007405281066895,
|
| 2261 |
+
"learning_rate": 7.871279248686794e-06,
|
| 2262 |
+
"loss": 0.1165,
|
| 2263 |
+
"step": 145000
|
| 2264 |
+
},
|
| 2265 |
+
{
|
| 2266 |
+
"epoch": 5.786208542114053,
|
| 2267 |
+
"grad_norm": 1.6658598184585571,
|
| 2268 |
+
"learning_rate": 7.2080437204860195e-06,
|
| 2269 |
+
"loss": 0.1171,
|
| 2270 |
+
"step": 145500
|
| 2271 |
+
},
|
| 2272 |
+
{
|
| 2273 |
+
"epoch": 5.800922611946234,
|
| 2274 |
+
"eval_loss": 0.21665571630001068,
|
| 2275 |
+
"eval_runtime": 31.2961,
|
| 2276 |
+
"eval_samples_per_second": 15.976,
|
| 2277 |
+
"eval_steps_per_second": 7.988,
|
| 2278 |
+
"step": 145870
|
| 2279 |
+
},
|
| 2280 |
+
{
|
| 2281 |
+
"epoch": 5.8060924202656485,
|
| 2282 |
+
"grad_norm": 3.7066237926483154,
|
| 2283 |
+
"learning_rate": 6.5461346633416465e-06,
|
| 2284 |
+
"loss": 0.1051,
|
| 2285 |
+
"step": 146000
|
| 2286 |
+
},
|
| 2287 |
+
{
|
| 2288 |
+
"epoch": 5.825976298417244,
|
| 2289 |
+
"grad_norm": 0.7491363286972046,
|
| 2290 |
+
"learning_rate": 5.882899135140872e-06,
|
| 2291 |
+
"loss": 0.1085,
|
| 2292 |
+
"step": 146500
|
| 2293 |
+
},
|
| 2294 |
+
{
|
| 2295 |
+
"epoch": 5.845860176568838,
|
| 2296 |
+
"grad_norm": 3.1486012935638428,
|
| 2297 |
+
"learning_rate": 5.219663606940097e-06,
|
| 2298 |
+
"loss": 0.1128,
|
| 2299 |
+
"step": 147000
|
| 2300 |
+
},
|
| 2301 |
+
{
|
| 2302 |
+
"epoch": 5.865744054720433,
|
| 2303 |
+
"grad_norm": 1.4818662405014038,
|
| 2304 |
+
"learning_rate": 4.556428078739322e-06,
|
| 2305 |
+
"loss": 0.1094,
|
| 2306 |
+
"step": 147500
|
| 2307 |
+
},
|
| 2308 |
+
{
|
| 2309 |
+
"epoch": 5.885627932872027,
|
| 2310 |
+
"grad_norm": 4.039377212524414,
|
| 2311 |
+
"learning_rate": 3.8931925505385475e-06,
|
| 2312 |
+
"loss": 0.1185,
|
| 2313 |
+
"step": 148000
|
| 2314 |
+
},
|
| 2315 |
+
{
|
| 2316 |
+
"epoch": 5.905511811023622,
|
| 2317 |
+
"grad_norm": 3.8260436058044434,
|
| 2318 |
+
"learning_rate": 3.2312834933941744e-06,
|
| 2319 |
+
"loss": 0.1053,
|
| 2320 |
+
"step": 148500
|
| 2321 |
+
},
|
| 2322 |
+
{
|
| 2323 |
+
"epoch": 5.925395689175216,
|
| 2324 |
+
"grad_norm": 2.552076578140259,
|
| 2325 |
+
"learning_rate": 2.5680479651933997e-06,
|
| 2326 |
+
"loss": 0.0979,
|
| 2327 |
+
"step": 149000
|
| 2328 |
+
},
|
| 2329 |
+
{
|
| 2330 |
+
"epoch": 5.9452795673268115,
|
| 2331 |
+
"grad_norm": 1.905666708946228,
|
| 2332 |
+
"learning_rate": 1.904812436992625e-06,
|
| 2333 |
+
"loss": 0.1081,
|
| 2334 |
+
"step": 149500
|
| 2335 |
+
},
|
| 2336 |
+
{
|
| 2337 |
+
"epoch": 5.965163445478407,
|
| 2338 |
+
"grad_norm": 4.959332466125488,
|
| 2339 |
+
"learning_rate": 1.2415769087918502e-06,
|
| 2340 |
+
"loss": 0.11,
|
| 2341 |
+
"step": 150000
|
| 2342 |
+
},
|
| 2343 |
+
{
|
| 2344 |
+
"epoch": 5.985047323630001,
|
| 2345 |
+
"grad_norm": 4.18074893951416,
|
| 2346 |
+
"learning_rate": 5.783413805910756e-07,
|
| 2347 |
+
"loss": 0.1105,
|
| 2348 |
+
"step": 150500
|
| 2349 |
+
},
|
| 2350 |
+
{
|
| 2351 |
+
"epoch": 6.0,
|
| 2352 |
+
"step": 150876,
|
| 2353 |
+
"total_flos": 9.265867688492237e+17,
|
| 2354 |
+
"train_loss": 0.2175943843949059,
|
| 2355 |
+
"train_runtime": 77334.2624,
|
| 2356 |
+
"train_samples_per_second": 3.902,
|
| 2357 |
+
"train_steps_per_second": 1.951
|
| 2358 |
+
}
|
| 2359 |
+
],
|
| 2360 |
+
"logging_steps": 500,
|
| 2361 |
+
"max_steps": 150876,
|
| 2362 |
+
"num_input_tokens_seen": 0,
|
| 2363 |
+
"num_train_epochs": 6,
|
| 2364 |
+
"save_steps": 500,
|
| 2365 |
+
"stateful_callbacks": {
|
| 2366 |
+
"TrainerControl": {
|
| 2367 |
+
"args": {
|
| 2368 |
+
"should_epoch_stop": false,
|
| 2369 |
+
"should_evaluate": false,
|
| 2370 |
+
"should_log": false,
|
| 2371 |
+
"should_save": true,
|
| 2372 |
+
"should_training_stop": true
|
| 2373 |
+
},
|
| 2374 |
+
"attributes": {}
|
| 2375 |
+
}
|
| 2376 |
+
},
|
| 2377 |
+
"total_flos": 9.265867688492237e+17,
|
| 2378 |
+
"train_batch_size": 2,
|
| 2379 |
+
"trial_name": null,
|
| 2380 |
+
"trial_params": null
|
| 2381 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4e3585f37a8d2c96562c7b0efa8fd397315640a822d7568f29981fbadebf7e5d
|
| 3 |
+
size 5496
|