End of training
Browse files- README.md +89 -0
- adapter_config.json +32 -0
- adapter_model.safetensors +3 -0
- loss_plot.png +0 -0
- trainer_state.json +1856 -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-hol-thms-by-file-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-hol-thms-by-file-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: 0.2839
|
| 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: 16
|
| 40 |
+
- eval_batch_size: 16
|
| 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 |
+
| 0.4174 | 0.2000 | 3777 | 0.4078 |
|
| 53 |
+
| 0.34 | 0.4000 | 7554 | 0.3589 |
|
| 54 |
+
| 0.3104 | 0.6001 | 11331 | 0.3305 |
|
| 55 |
+
| 0.2901 | 0.8001 | 15108 | 0.3113 |
|
| 56 |
+
| 0.2823 | 1.0001 | 18885 | 0.2998 |
|
| 57 |
+
| 0.2591 | 1.2001 | 22662 | 0.2958 |
|
| 58 |
+
| 0.2597 | 1.4001 | 26439 | 0.2871 |
|
| 59 |
+
| 0.2536 | 1.6002 | 30216 | 0.2895 |
|
| 60 |
+
| 0.2353 | 1.8002 | 33993 | 0.2819 |
|
| 61 |
+
| 0.2236 | 2.0002 | 37770 | 0.2784 |
|
| 62 |
+
| 0.2055 | 2.2002 | 41547 | 0.2898 |
|
| 63 |
+
| 0.2129 | 2.4003 | 45324 | 0.2687 |
|
| 64 |
+
| 0.2001 | 2.6003 | 49101 | 0.2719 |
|
| 65 |
+
| 0.2108 | 2.8003 | 52878 | 0.2714 |
|
| 66 |
+
| 0.2023 | 3.0003 | 56655 | 0.2650 |
|
| 67 |
+
| 0.1811 | 3.2003 | 60432 | 0.2709 |
|
| 68 |
+
| 0.1702 | 3.4004 | 64209 | 0.2655 |
|
| 69 |
+
| 0.176 | 3.6004 | 67986 | 0.2665 |
|
| 70 |
+
| 0.1702 | 3.8004 | 71763 | 0.2666 |
|
| 71 |
+
| 0.1597 | 4.0004 | 75540 | 0.2620 |
|
| 72 |
+
| 0.1388 | 4.2004 | 79317 | 0.2704 |
|
| 73 |
+
| 0.1452 | 4.4005 | 83094 | 0.2707 |
|
| 74 |
+
| 0.146 | 4.6005 | 86871 | 0.2705 |
|
| 75 |
+
| 0.1388 | 4.8005 | 90648 | 0.2622 |
|
| 76 |
+
| 0.1435 | 5.0005 | 94425 | 0.2649 |
|
| 77 |
+
| 0.1295 | 5.2006 | 98202 | 0.2794 |
|
| 78 |
+
| 0.1195 | 5.4006 | 101979 | 0.2780 |
|
| 79 |
+
| 0.124 | 5.6006 | 105756 | 0.2796 |
|
| 80 |
+
| 0.1183 | 5.8006 | 109533 | 0.2839 |
|
| 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 sha256:57492a918167b88f865fdeb7cdab588ab1b036cf6ab7bbe4ff32b788df404eda
|
| 3 |
+
size 531035104
|
loss_plot.png
ADDED
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1856 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 6.0,
|
| 5 |
+
"eval_steps": 3777,
|
| 6 |
+
"global_step": 113298,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.026478843404120107,
|
| 13 |
+
"grad_norm": 2.1454062461853027,
|
| 14 |
+
"learning_rate": 0.00019929327373275146,
|
| 15 |
+
"loss": 0.8356,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.052957686808240215,
|
| 20 |
+
"grad_norm": 1.2584812641143799,
|
| 21 |
+
"learning_rate": 0.00019841163271435891,
|
| 22 |
+
"loss": 0.5745,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.07943653021236033,
|
| 27 |
+
"grad_norm": 1.0932960510253906,
|
| 28 |
+
"learning_rate": 0.00019752822488029826,
|
| 29 |
+
"loss": 0.5041,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.10591537361648043,
|
| 34 |
+
"grad_norm": 2.201864004135132,
|
| 35 |
+
"learning_rate": 0.00019664481704623757,
|
| 36 |
+
"loss": 0.4696,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.13239421702060053,
|
| 41 |
+
"grad_norm": 1.9619899988174438,
|
| 42 |
+
"learning_rate": 0.0001957614092121769,
|
| 43 |
+
"loss": 0.4524,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.15887306042472066,
|
| 48 |
+
"grad_norm": 2.994553327560425,
|
| 49 |
+
"learning_rate": 0.00019487800137811624,
|
| 50 |
+
"loss": 0.418,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.18535190382884076,
|
| 55 |
+
"grad_norm": 2.0418312549591064,
|
| 56 |
+
"learning_rate": 0.00019399459354405555,
|
| 57 |
+
"loss": 0.4174,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.2000211830747233,
|
| 62 |
+
"eval_loss": 0.40778520703315735,
|
| 63 |
+
"eval_runtime": 421.2687,
|
| 64 |
+
"eval_samples_per_second": 10.936,
|
| 65 |
+
"eval_steps_per_second": 5.469,
|
| 66 |
+
"step": 3777
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.21183074723296086,
|
| 70 |
+
"grad_norm": 1.8520241975784302,
|
| 71 |
+
"learning_rate": 0.0001931111857099949,
|
| 72 |
+
"loss": 0.4004,
|
| 73 |
+
"step": 4000
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.23830959063708096,
|
| 77 |
+
"grad_norm": 1.4545440673828125,
|
| 78 |
+
"learning_rate": 0.00019222954469160232,
|
| 79 |
+
"loss": 0.3804,
|
| 80 |
+
"step": 4500
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.26478843404120106,
|
| 84 |
+
"grad_norm": 2.6964781284332275,
|
| 85 |
+
"learning_rate": 0.00019134613685754166,
|
| 86 |
+
"loss": 0.3777,
|
| 87 |
+
"step": 5000
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.2912672774453212,
|
| 91 |
+
"grad_norm": 2.227689027786255,
|
| 92 |
+
"learning_rate": 0.00019046272902348098,
|
| 93 |
+
"loss": 0.3642,
|
| 94 |
+
"step": 5500
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.3177461208494413,
|
| 98 |
+
"grad_norm": 2.4446637630462646,
|
| 99 |
+
"learning_rate": 0.00018957932118942032,
|
| 100 |
+
"loss": 0.359,
|
| 101 |
+
"step": 6000
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.3442249642535614,
|
| 105 |
+
"grad_norm": 1.461595892906189,
|
| 106 |
+
"learning_rate": 0.00018869591335535964,
|
| 107 |
+
"loss": 0.3509,
|
| 108 |
+
"step": 6500
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.3707038076576815,
|
| 112 |
+
"grad_norm": 1.95978581905365,
|
| 113 |
+
"learning_rate": 0.00018781250552129896,
|
| 114 |
+
"loss": 0.3561,
|
| 115 |
+
"step": 7000
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.3971826510618016,
|
| 119 |
+
"grad_norm": 2.1562278270721436,
|
| 120 |
+
"learning_rate": 0.0001869290976872383,
|
| 121 |
+
"loss": 0.34,
|
| 122 |
+
"step": 7500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.4000423661494466,
|
| 126 |
+
"eval_loss": 0.35886627435684204,
|
| 127 |
+
"eval_runtime": 421.2564,
|
| 128 |
+
"eval_samples_per_second": 10.936,
|
| 129 |
+
"eval_steps_per_second": 5.469,
|
| 130 |
+
"step": 7554
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 0.4236614944659217,
|
| 134 |
+
"grad_norm": 2.0626142024993896,
|
| 135 |
+
"learning_rate": 0.00018604745666884575,
|
| 136 |
+
"loss": 0.3374,
|
| 137 |
+
"step": 8000
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 0.4501403378700418,
|
| 141 |
+
"grad_norm": 1.5433763265609741,
|
| 142 |
+
"learning_rate": 0.00018516581565045318,
|
| 143 |
+
"loss": 0.3371,
|
| 144 |
+
"step": 8500
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 0.4766191812741619,
|
| 148 |
+
"grad_norm": 2.220177412033081,
|
| 149 |
+
"learning_rate": 0.00018428240781639252,
|
| 150 |
+
"loss": 0.3325,
|
| 151 |
+
"step": 9000
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 0.503098024678282,
|
| 155 |
+
"grad_norm": 1.1187978982925415,
|
| 156 |
+
"learning_rate": 0.00018339899998233187,
|
| 157 |
+
"loss": 0.3245,
|
| 158 |
+
"step": 9500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.5295768680824021,
|
| 162 |
+
"grad_norm": 2.554443836212158,
|
| 163 |
+
"learning_rate": 0.00018251559214827118,
|
| 164 |
+
"loss": 0.3252,
|
| 165 |
+
"step": 10000
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.5560557114865222,
|
| 169 |
+
"grad_norm": 1.8116191625595093,
|
| 170 |
+
"learning_rate": 0.0001816321843142105,
|
| 171 |
+
"loss": 0.3153,
|
| 172 |
+
"step": 10500
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.5825345548906424,
|
| 176 |
+
"grad_norm": 1.6784979104995728,
|
| 177 |
+
"learning_rate": 0.00018074877648014982,
|
| 178 |
+
"loss": 0.3104,
|
| 179 |
+
"step": 11000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.6000635492241699,
|
| 183 |
+
"eval_loss": 0.3305271863937378,
|
| 184 |
+
"eval_runtime": 421.3809,
|
| 185 |
+
"eval_samples_per_second": 10.933,
|
| 186 |
+
"eval_steps_per_second": 5.468,
|
| 187 |
+
"step": 11331
|
| 188 |
+
},
|
| 189 |
+
{
|
| 190 |
+
"epoch": 0.6090133982947625,
|
| 191 |
+
"grad_norm": 2.8816235065460205,
|
| 192 |
+
"learning_rate": 0.00017986536864608916,
|
| 193 |
+
"loss": 0.3211,
|
| 194 |
+
"step": 11500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.6354922416988826,
|
| 198 |
+
"grad_norm": 0.8169271945953369,
|
| 199 |
+
"learning_rate": 0.0001789819608120285,
|
| 200 |
+
"loss": 0.3172,
|
| 201 |
+
"step": 12000
|
| 202 |
+
},
|
| 203 |
+
{
|
| 204 |
+
"epoch": 0.6619710851030027,
|
| 205 |
+
"grad_norm": 1.5912612676620483,
|
| 206 |
+
"learning_rate": 0.00017809855297796782,
|
| 207 |
+
"loss": 0.3092,
|
| 208 |
+
"step": 12500
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 0.6884499285071228,
|
| 212 |
+
"grad_norm": 1.9082287549972534,
|
| 213 |
+
"learning_rate": 0.00017721691195957527,
|
| 214 |
+
"loss": 0.302,
|
| 215 |
+
"step": 13000
|
| 216 |
+
},
|
| 217 |
+
{
|
| 218 |
+
"epoch": 0.7149287719112429,
|
| 219 |
+
"grad_norm": 1.7852967977523804,
|
| 220 |
+
"learning_rate": 0.0001763335041255146,
|
| 221 |
+
"loss": 0.3026,
|
| 222 |
+
"step": 13500
|
| 223 |
+
},
|
| 224 |
+
{
|
| 225 |
+
"epoch": 0.741407615315363,
|
| 226 |
+
"grad_norm": 1.7745851278305054,
|
| 227 |
+
"learning_rate": 0.00017545009629145393,
|
| 228 |
+
"loss": 0.2926,
|
| 229 |
+
"step": 14000
|
| 230 |
+
},
|
| 231 |
+
{
|
| 232 |
+
"epoch": 0.7678864587194831,
|
| 233 |
+
"grad_norm": 1.4642239809036255,
|
| 234 |
+
"learning_rate": 0.00017456668845739325,
|
| 235 |
+
"loss": 0.2987,
|
| 236 |
+
"step": 14500
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.7943653021236032,
|
| 240 |
+
"grad_norm": 1.8523545265197754,
|
| 241 |
+
"learning_rate": 0.00017368328062333257,
|
| 242 |
+
"loss": 0.2901,
|
| 243 |
+
"step": 15000
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.8000847322988932,
|
| 247 |
+
"eval_loss": 0.3112636208534241,
|
| 248 |
+
"eval_runtime": 420.9037,
|
| 249 |
+
"eval_samples_per_second": 10.945,
|
| 250 |
+
"eval_steps_per_second": 5.474,
|
| 251 |
+
"step": 15108
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 0.8208441455277233,
|
| 255 |
+
"grad_norm": 3.936985492706299,
|
| 256 |
+
"learning_rate": 0.00017280163960494002,
|
| 257 |
+
"loss": 0.289,
|
| 258 |
+
"step": 15500
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 0.8473229889318434,
|
| 262 |
+
"grad_norm": 2.120314598083496,
|
| 263 |
+
"learning_rate": 0.00017191823177087936,
|
| 264 |
+
"loss": 0.2947,
|
| 265 |
+
"step": 16000
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 0.8738018323359635,
|
| 269 |
+
"grad_norm": 2.1625630855560303,
|
| 270 |
+
"learning_rate": 0.00017103659075248678,
|
| 271 |
+
"loss": 0.2897,
|
| 272 |
+
"step": 16500
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.9002806757400836,
|
| 276 |
+
"grad_norm": 1.2618048191070557,
|
| 277 |
+
"learning_rate": 0.00017015318291842613,
|
| 278 |
+
"loss": 0.2919,
|
| 279 |
+
"step": 17000
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 0.9267595191442037,
|
| 283 |
+
"grad_norm": 1.7789500951766968,
|
| 284 |
+
"learning_rate": 0.00016926977508436545,
|
| 285 |
+
"loss": 0.2841,
|
| 286 |
+
"step": 17500
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 0.9532383625483238,
|
| 290 |
+
"grad_norm": 0.8625043034553528,
|
| 291 |
+
"learning_rate": 0.0001683863672503048,
|
| 292 |
+
"loss": 0.2813,
|
| 293 |
+
"step": 18000
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 0.979717205952444,
|
| 297 |
+
"grad_norm": 2.622668981552124,
|
| 298 |
+
"learning_rate": 0.00016750295941624413,
|
| 299 |
+
"loss": 0.2823,
|
| 300 |
+
"step": 18500
|
| 301 |
+
},
|
| 302 |
+
{
|
| 303 |
+
"epoch": 1.0001059153736165,
|
| 304 |
+
"eval_loss": 0.2998111844062805,
|
| 305 |
+
"eval_runtime": 420.9652,
|
| 306 |
+
"eval_samples_per_second": 10.944,
|
| 307 |
+
"eval_steps_per_second": 5.473,
|
| 308 |
+
"step": 18885
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 1.006196049356564,
|
| 312 |
+
"grad_norm": 1.9936221837997437,
|
| 313 |
+
"learning_rate": 0.00016661955158218342,
|
| 314 |
+
"loss": 0.2725,
|
| 315 |
+
"step": 19000
|
| 316 |
+
},
|
| 317 |
+
{
|
| 318 |
+
"epoch": 1.0326748927606841,
|
| 319 |
+
"grad_norm": 1.7693616151809692,
|
| 320 |
+
"learning_rate": 0.00016573614374812277,
|
| 321 |
+
"loss": 0.2572,
|
| 322 |
+
"step": 19500
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 1.0591537361648042,
|
| 326 |
+
"grad_norm": 1.0694758892059326,
|
| 327 |
+
"learning_rate": 0.00016485273591406208,
|
| 328 |
+
"loss": 0.2526,
|
| 329 |
+
"step": 20000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 1.0856325795689243,
|
| 333 |
+
"grad_norm": 2.6468887329101562,
|
| 334 |
+
"learning_rate": 0.00016396932808000143,
|
| 335 |
+
"loss": 0.257,
|
| 336 |
+
"step": 20500
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 1.1121114229730444,
|
| 340 |
+
"grad_norm": 2.7952466011047363,
|
| 341 |
+
"learning_rate": 0.00016308768706160885,
|
| 342 |
+
"loss": 0.2583,
|
| 343 |
+
"step": 21000
|
| 344 |
+
},
|
| 345 |
+
{
|
| 346 |
+
"epoch": 1.1385902663771645,
|
| 347 |
+
"grad_norm": 4.5573625564575195,
|
| 348 |
+
"learning_rate": 0.0001622042792275482,
|
| 349 |
+
"loss": 0.2601,
|
| 350 |
+
"step": 21500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 1.1650691097812849,
|
| 354 |
+
"grad_norm": 1.8404908180236816,
|
| 355 |
+
"learning_rate": 0.00016132087139348754,
|
| 356 |
+
"loss": 0.2475,
|
| 357 |
+
"step": 22000
|
| 358 |
+
},
|
| 359 |
+
{
|
| 360 |
+
"epoch": 1.1915479531854047,
|
| 361 |
+
"grad_norm": 1.9419597387313843,
|
| 362 |
+
"learning_rate": 0.00016043746355942686,
|
| 363 |
+
"loss": 0.2591,
|
| 364 |
+
"step": 22500
|
| 365 |
+
},
|
| 366 |
+
{
|
| 367 |
+
"epoch": 1.2001270984483399,
|
| 368 |
+
"eval_loss": 0.29582297801971436,
|
| 369 |
+
"eval_runtime": 421.9827,
|
| 370 |
+
"eval_samples_per_second": 10.918,
|
| 371 |
+
"eval_steps_per_second": 5.46,
|
| 372 |
+
"step": 22662
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 1.218026796589525,
|
| 376 |
+
"grad_norm": 3.7483320236206055,
|
| 377 |
+
"learning_rate": 0.00015955405572536617,
|
| 378 |
+
"loss": 0.2438,
|
| 379 |
+
"step": 23000
|
| 380 |
+
},
|
| 381 |
+
{
|
| 382 |
+
"epoch": 1.2445056399936452,
|
| 383 |
+
"grad_norm": 1.0397343635559082,
|
| 384 |
+
"learning_rate": 0.00015867418152264176,
|
| 385 |
+
"loss": 0.2546,
|
| 386 |
+
"step": 23500
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 1.2709844833977653,
|
| 390 |
+
"grad_norm": 1.655588150024414,
|
| 391 |
+
"learning_rate": 0.00015779077368858108,
|
| 392 |
+
"loss": 0.2517,
|
| 393 |
+
"step": 24000
|
| 394 |
+
},
|
| 395 |
+
{
|
| 396 |
+
"epoch": 1.2974633268018854,
|
| 397 |
+
"grad_norm": 3.5589778423309326,
|
| 398 |
+
"learning_rate": 0.00015690736585452042,
|
| 399 |
+
"loss": 0.2503,
|
| 400 |
+
"step": 24500
|
| 401 |
+
},
|
| 402 |
+
{
|
| 403 |
+
"epoch": 1.3239421702060055,
|
| 404 |
+
"grad_norm": 1.4428561925888062,
|
| 405 |
+
"learning_rate": 0.00015602395802045974,
|
| 406 |
+
"loss": 0.2521,
|
| 407 |
+
"step": 25000
|
| 408 |
+
},
|
| 409 |
+
{
|
| 410 |
+
"epoch": 1.3504210136101256,
|
| 411 |
+
"grad_norm": 1.6518936157226562,
|
| 412 |
+
"learning_rate": 0.00015514055018639905,
|
| 413 |
+
"loss": 0.2513,
|
| 414 |
+
"step": 25500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 1.3768998570142457,
|
| 418 |
+
"grad_norm": 1.4650779962539673,
|
| 419 |
+
"learning_rate": 0.0001542571423523384,
|
| 420 |
+
"loss": 0.2597,
|
| 421 |
+
"step": 26000
|
| 422 |
+
},
|
| 423 |
+
{
|
| 424 |
+
"epoch": 1.400148281523063,
|
| 425 |
+
"eval_loss": 0.2870725393295288,
|
| 426 |
+
"eval_runtime": 421.6939,
|
| 427 |
+
"eval_samples_per_second": 10.925,
|
| 428 |
+
"eval_steps_per_second": 5.464,
|
| 429 |
+
"step": 26439
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 1.4033787004183658,
|
| 433 |
+
"grad_norm": 2.4973056316375732,
|
| 434 |
+
"learning_rate": 0.0001533737345182777,
|
| 435 |
+
"loss": 0.2438,
|
| 436 |
+
"step": 26500
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 1.4298575438224859,
|
| 440 |
+
"grad_norm": 1.4028382301330566,
|
| 441 |
+
"learning_rate": 0.00015249032668421703,
|
| 442 |
+
"loss": 0.2414,
|
| 443 |
+
"step": 27000
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 1.456336387226606,
|
| 447 |
+
"grad_norm": 1.586595058441162,
|
| 448 |
+
"learning_rate": 0.00015160691885015637,
|
| 449 |
+
"loss": 0.245,
|
| 450 |
+
"step": 27500
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 1.482815230630726,
|
| 454 |
+
"grad_norm": 2.191901922225952,
|
| 455 |
+
"learning_rate": 0.00015072527783176383,
|
| 456 |
+
"loss": 0.246,
|
| 457 |
+
"step": 28000
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 1.5092940740348462,
|
| 461 |
+
"grad_norm": 1.2381248474121094,
|
| 462 |
+
"learning_rate": 0.00014984186999770314,
|
| 463 |
+
"loss": 0.2449,
|
| 464 |
+
"step": 28500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 1.5357729174389663,
|
| 468 |
+
"grad_norm": 1.502170443534851,
|
| 469 |
+
"learning_rate": 0.00014895846216364246,
|
| 470 |
+
"loss": 0.2448,
|
| 471 |
+
"step": 29000
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 1.5622517608430864,
|
| 475 |
+
"grad_norm": 2.206437349319458,
|
| 476 |
+
"learning_rate": 0.0001480768211452499,
|
| 477 |
+
"loss": 0.2488,
|
| 478 |
+
"step": 29500
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.5887306042472065,
|
| 482 |
+
"grad_norm": 2.766979694366455,
|
| 483 |
+
"learning_rate": 0.00014719341331118925,
|
| 484 |
+
"loss": 0.2536,
|
| 485 |
+
"step": 30000
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.6001694645977864,
|
| 489 |
+
"eval_loss": 0.2894779145717621,
|
| 490 |
+
"eval_runtime": 421.1711,
|
| 491 |
+
"eval_samples_per_second": 10.939,
|
| 492 |
+
"eval_steps_per_second": 5.47,
|
| 493 |
+
"step": 30216
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 1.6152094476513266,
|
| 497 |
+
"grad_norm": 1.9961110353469849,
|
| 498 |
+
"learning_rate": 0.0001463100054771286,
|
| 499 |
+
"loss": 0.2312,
|
| 500 |
+
"step": 30500
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 1.6416882910554467,
|
| 504 |
+
"grad_norm": 3.271880626678467,
|
| 505 |
+
"learning_rate": 0.0001454265976430679,
|
| 506 |
+
"loss": 0.2387,
|
| 507 |
+
"step": 31000
|
| 508 |
+
},
|
| 509 |
+
{
|
| 510 |
+
"epoch": 1.6681671344595668,
|
| 511 |
+
"grad_norm": 0.7093228697776794,
|
| 512 |
+
"learning_rate": 0.00014454318980900723,
|
| 513 |
+
"loss": 0.239,
|
| 514 |
+
"step": 31500
|
| 515 |
+
},
|
| 516 |
+
{
|
| 517 |
+
"epoch": 1.6946459778636869,
|
| 518 |
+
"grad_norm": 1.9597697257995605,
|
| 519 |
+
"learning_rate": 0.00014366154879061468,
|
| 520 |
+
"loss": 0.2335,
|
| 521 |
+
"step": 32000
|
| 522 |
+
},
|
| 523 |
+
{
|
| 524 |
+
"epoch": 1.721124821267807,
|
| 525 |
+
"grad_norm": 2.2032573223114014,
|
| 526 |
+
"learning_rate": 0.00014277814095655403,
|
| 527 |
+
"loss": 0.2399,
|
| 528 |
+
"step": 32500
|
| 529 |
+
},
|
| 530 |
+
{
|
| 531 |
+
"epoch": 1.747603664671927,
|
| 532 |
+
"grad_norm": 1.6669458150863647,
|
| 533 |
+
"learning_rate": 0.00014189473312249332,
|
| 534 |
+
"loss": 0.2472,
|
| 535 |
+
"step": 33000
|
| 536 |
+
},
|
| 537 |
+
{
|
| 538 |
+
"epoch": 1.7740825080760474,
|
| 539 |
+
"grad_norm": 2.5589282512664795,
|
| 540 |
+
"learning_rate": 0.00014101132528843266,
|
| 541 |
+
"loss": 0.2353,
|
| 542 |
+
"step": 33500
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 1.8001906476725096,
|
| 546 |
+
"eval_loss": 0.2819114029407501,
|
| 547 |
+
"eval_runtime": 420.9242,
|
| 548 |
+
"eval_samples_per_second": 10.945,
|
| 549 |
+
"eval_steps_per_second": 5.474,
|
| 550 |
+
"step": 33993
|
| 551 |
+
},
|
| 552 |
+
{
|
| 553 |
+
"epoch": 1.8005613514801673,
|
| 554 |
+
"grad_norm": 2.796818971633911,
|
| 555 |
+
"learning_rate": 0.000140127917454372,
|
| 556 |
+
"loss": 0.2399,
|
| 557 |
+
"step": 34000
|
| 558 |
+
},
|
| 559 |
+
{
|
| 560 |
+
"epoch": 1.8270401948842876,
|
| 561 |
+
"grad_norm": 1.920599341392517,
|
| 562 |
+
"learning_rate": 0.00013924450962031132,
|
| 563 |
+
"loss": 0.2363,
|
| 564 |
+
"step": 34500
|
| 565 |
+
},
|
| 566 |
+
{
|
| 567 |
+
"epoch": 1.8535190382884075,
|
| 568 |
+
"grad_norm": 1.8848789930343628,
|
| 569 |
+
"learning_rate": 0.00013836110178625066,
|
| 570 |
+
"loss": 0.2357,
|
| 571 |
+
"step": 35000
|
| 572 |
+
},
|
| 573 |
+
{
|
| 574 |
+
"epoch": 1.8799978816925278,
|
| 575 |
+
"grad_norm": 2.0130763053894043,
|
| 576 |
+
"learning_rate": 0.00013747769395218998,
|
| 577 |
+
"loss": 0.2393,
|
| 578 |
+
"step": 35500
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 1.9064767250966477,
|
| 582 |
+
"grad_norm": 3.166252613067627,
|
| 583 |
+
"learning_rate": 0.00013659605293379743,
|
| 584 |
+
"loss": 0.2369,
|
| 585 |
+
"step": 36000
|
| 586 |
+
},
|
| 587 |
+
{
|
| 588 |
+
"epoch": 1.932955568500768,
|
| 589 |
+
"grad_norm": 2.393334150314331,
|
| 590 |
+
"learning_rate": 0.00013571264509973675,
|
| 591 |
+
"loss": 0.2383,
|
| 592 |
+
"step": 36500
|
| 593 |
+
},
|
| 594 |
+
{
|
| 595 |
+
"epoch": 1.9594344119048879,
|
| 596 |
+
"grad_norm": 1.5478434562683105,
|
| 597 |
+
"learning_rate": 0.0001348292372656761,
|
| 598 |
+
"loss": 0.2385,
|
| 599 |
+
"step": 37000
|
| 600 |
+
},
|
| 601 |
+
{
|
| 602 |
+
"epoch": 1.9859132553090082,
|
| 603 |
+
"grad_norm": 1.8537359237670898,
|
| 604 |
+
"learning_rate": 0.0001339458294316154,
|
| 605 |
+
"loss": 0.2236,
|
| 606 |
+
"step": 37500
|
| 607 |
+
},
|
| 608 |
+
{
|
| 609 |
+
"epoch": 2.000211830747233,
|
| 610 |
+
"eval_loss": 0.278350293636322,
|
| 611 |
+
"eval_runtime": 420.507,
|
| 612 |
+
"eval_samples_per_second": 10.956,
|
| 613 |
+
"eval_steps_per_second": 5.479,
|
| 614 |
+
"step": 37770
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 2.012392098713128,
|
| 618 |
+
"grad_norm": 1.4909721612930298,
|
| 619 |
+
"learning_rate": 0.00013306242159755473,
|
| 620 |
+
"loss": 0.2094,
|
| 621 |
+
"step": 38000
|
| 622 |
+
},
|
| 623 |
+
{
|
| 624 |
+
"epoch": 2.0388709421172484,
|
| 625 |
+
"grad_norm": 4.634109020233154,
|
| 626 |
+
"learning_rate": 0.00013217901376349407,
|
| 627 |
+
"loss": 0.2012,
|
| 628 |
+
"step": 38500
|
| 629 |
+
},
|
| 630 |
+
{
|
| 631 |
+
"epoch": 2.0653497855213683,
|
| 632 |
+
"grad_norm": 2.142557382583618,
|
| 633 |
+
"learning_rate": 0.0001312973727451015,
|
| 634 |
+
"loss": 0.2084,
|
| 635 |
+
"step": 39000
|
| 636 |
+
},
|
| 637 |
+
{
|
| 638 |
+
"epoch": 2.0918286289254886,
|
| 639 |
+
"grad_norm": 4.682730674743652,
|
| 640 |
+
"learning_rate": 0.00013041396491104084,
|
| 641 |
+
"loss": 0.2089,
|
| 642 |
+
"step": 39500
|
| 643 |
+
},
|
| 644 |
+
{
|
| 645 |
+
"epoch": 2.1183074723296085,
|
| 646 |
+
"grad_norm": 2.030521869659424,
|
| 647 |
+
"learning_rate": 0.0001295323238926483,
|
| 648 |
+
"loss": 0.2109,
|
| 649 |
+
"step": 40000
|
| 650 |
+
},
|
| 651 |
+
{
|
| 652 |
+
"epoch": 2.144786315733729,
|
| 653 |
+
"grad_norm": 2.6542880535125732,
|
| 654 |
+
"learning_rate": 0.00012864891605858763,
|
| 655 |
+
"loss": 0.2,
|
| 656 |
+
"step": 40500
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 2.1712651591378487,
|
| 660 |
+
"grad_norm": 1.762758493423462,
|
| 661 |
+
"learning_rate": 0.00012776550822452692,
|
| 662 |
+
"loss": 0.2139,
|
| 663 |
+
"step": 41000
|
| 664 |
+
},
|
| 665 |
+
{
|
| 666 |
+
"epoch": 2.197744002541969,
|
| 667 |
+
"grad_norm": 0.5812512636184692,
|
| 668 |
+
"learning_rate": 0.00012688210039046627,
|
| 669 |
+
"loss": 0.2055,
|
| 670 |
+
"step": 41500
|
| 671 |
+
},
|
| 672 |
+
{
|
| 673 |
+
"epoch": 2.2002330138219564,
|
| 674 |
+
"eval_loss": 0.2898082733154297,
|
| 675 |
+
"eval_runtime": 421.4286,
|
| 676 |
+
"eval_samples_per_second": 10.932,
|
| 677 |
+
"eval_steps_per_second": 5.467,
|
| 678 |
+
"step": 41547
|
| 679 |
+
},
|
| 680 |
+
{
|
| 681 |
+
"epoch": 2.224222845946089,
|
| 682 |
+
"grad_norm": 1.5051369667053223,
|
| 683 |
+
"learning_rate": 0.00012599869255640558,
|
| 684 |
+
"loss": 0.2099,
|
| 685 |
+
"step": 42000
|
| 686 |
+
},
|
| 687 |
+
{
|
| 688 |
+
"epoch": 2.250701689350209,
|
| 689 |
+
"grad_norm": 2.4985218048095703,
|
| 690 |
+
"learning_rate": 0.00012511528472234493,
|
| 691 |
+
"loss": 0.2038,
|
| 692 |
+
"step": 42500
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 2.277180532754329,
|
| 696 |
+
"grad_norm": 3.212822914123535,
|
| 697 |
+
"learning_rate": 0.00012423187688828427,
|
| 698 |
+
"loss": 0.2008,
|
| 699 |
+
"step": 43000
|
| 700 |
+
},
|
| 701 |
+
{
|
| 702 |
+
"epoch": 2.3036593761584494,
|
| 703 |
+
"grad_norm": 1.2164196968078613,
|
| 704 |
+
"learning_rate": 0.00012334846905422356,
|
| 705 |
+
"loss": 0.2078,
|
| 706 |
+
"step": 43500
|
| 707 |
+
},
|
| 708 |
+
{
|
| 709 |
+
"epoch": 2.3301382195625697,
|
| 710 |
+
"grad_norm": 0.8281005024909973,
|
| 711 |
+
"learning_rate": 0.00012246682803583104,
|
| 712 |
+
"loss": 0.208,
|
| 713 |
+
"step": 44000
|
| 714 |
+
},
|
| 715 |
+
{
|
| 716 |
+
"epoch": 2.3566170629666896,
|
| 717 |
+
"grad_norm": 2.5285778045654297,
|
| 718 |
+
"learning_rate": 0.00012158342020177036,
|
| 719 |
+
"loss": 0.2105,
|
| 720 |
+
"step": 44500
|
| 721 |
+
},
|
| 722 |
+
{
|
| 723 |
+
"epoch": 2.3830959063708095,
|
| 724 |
+
"grad_norm": 0.9707619547843933,
|
| 725 |
+
"learning_rate": 0.00012070001236770969,
|
| 726 |
+
"loss": 0.2129,
|
| 727 |
+
"step": 45000
|
| 728 |
+
},
|
| 729 |
+
{
|
| 730 |
+
"epoch": 2.4002541968966797,
|
| 731 |
+
"eval_loss": 0.26870912313461304,
|
| 732 |
+
"eval_runtime": 421.9405,
|
| 733 |
+
"eval_samples_per_second": 10.919,
|
| 734 |
+
"eval_steps_per_second": 5.46,
|
| 735 |
+
"step": 45324
|
| 736 |
+
},
|
| 737 |
+
{
|
| 738 |
+
"epoch": 2.40957474977493,
|
| 739 |
+
"grad_norm": 1.2378835678100586,
|
| 740 |
+
"learning_rate": 0.000119816604533649,
|
| 741 |
+
"loss": 0.2072,
|
| 742 |
+
"step": 45500
|
| 743 |
+
},
|
| 744 |
+
{
|
| 745 |
+
"epoch": 2.43605359317905,
|
| 746 |
+
"grad_norm": 1.5940284729003906,
|
| 747 |
+
"learning_rate": 0.00011893319669958833,
|
| 748 |
+
"loss": 0.2031,
|
| 749 |
+
"step": 46000
|
| 750 |
+
},
|
| 751 |
+
{
|
| 752 |
+
"epoch": 2.46253243658317,
|
| 753 |
+
"grad_norm": 2.2567298412323,
|
| 754 |
+
"learning_rate": 0.00011804978886552766,
|
| 755 |
+
"loss": 0.205,
|
| 756 |
+
"step": 46500
|
| 757 |
+
},
|
| 758 |
+
{
|
| 759 |
+
"epoch": 2.4890112799872903,
|
| 760 |
+
"grad_norm": 1.5430265665054321,
|
| 761 |
+
"learning_rate": 0.000117166381031467,
|
| 762 |
+
"loss": 0.2041,
|
| 763 |
+
"step": 47000
|
| 764 |
+
},
|
| 765 |
+
{
|
| 766 |
+
"epoch": 2.51549012339141,
|
| 767 |
+
"grad_norm": 2.6054415702819824,
|
| 768 |
+
"learning_rate": 0.00011628297319740632,
|
| 769 |
+
"loss": 0.2055,
|
| 770 |
+
"step": 47500
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 2.5419689667955305,
|
| 774 |
+
"grad_norm": 1.0687452554702759,
|
| 775 |
+
"learning_rate": 0.00011540133217901376,
|
| 776 |
+
"loss": 0.2032,
|
| 777 |
+
"step": 48000
|
| 778 |
+
},
|
| 779 |
+
{
|
| 780 |
+
"epoch": 2.5684478101996504,
|
| 781 |
+
"grad_norm": 1.7363426685333252,
|
| 782 |
+
"learning_rate": 0.00011451792434495309,
|
| 783 |
+
"loss": 0.1997,
|
| 784 |
+
"step": 48500
|
| 785 |
+
},
|
| 786 |
+
{
|
| 787 |
+
"epoch": 2.5949266536037707,
|
| 788 |
+
"grad_norm": 1.5278480052947998,
|
| 789 |
+
"learning_rate": 0.00011363451651089244,
|
| 790 |
+
"loss": 0.2001,
|
| 791 |
+
"step": 49000
|
| 792 |
+
},
|
| 793 |
+
{
|
| 794 |
+
"epoch": 2.6002753799714027,
|
| 795 |
+
"eval_loss": 0.2719252407550812,
|
| 796 |
+
"eval_runtime": 420.6405,
|
| 797 |
+
"eval_samples_per_second": 10.952,
|
| 798 |
+
"eval_steps_per_second": 5.477,
|
| 799 |
+
"step": 49101
|
| 800 |
+
},
|
| 801 |
+
{
|
| 802 |
+
"epoch": 2.6214054970078906,
|
| 803 |
+
"grad_norm": 5.551360130310059,
|
| 804 |
+
"learning_rate": 0.00011275110867683177,
|
| 805 |
+
"loss": 0.2067,
|
| 806 |
+
"step": 49500
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 2.647884340412011,
|
| 810 |
+
"grad_norm": 0.9148823618888855,
|
| 811 |
+
"learning_rate": 0.00011186946765843919,
|
| 812 |
+
"loss": 0.202,
|
| 813 |
+
"step": 50000
|
| 814 |
+
},
|
| 815 |
+
{
|
| 816 |
+
"epoch": 2.674363183816131,
|
| 817 |
+
"grad_norm": 2.075822591781616,
|
| 818 |
+
"learning_rate": 0.00011098605982437854,
|
| 819 |
+
"loss": 0.1996,
|
| 820 |
+
"step": 50500
|
| 821 |
+
},
|
| 822 |
+
{
|
| 823 |
+
"epoch": 2.700842027220251,
|
| 824 |
+
"grad_norm": 4.498257160186768,
|
| 825 |
+
"learning_rate": 0.00011010441880598599,
|
| 826 |
+
"loss": 0.1908,
|
| 827 |
+
"step": 51000
|
| 828 |
+
},
|
| 829 |
+
{
|
| 830 |
+
"epoch": 2.727320870624371,
|
| 831 |
+
"grad_norm": 3.7217941284179688,
|
| 832 |
+
"learning_rate": 0.0001092210109719253,
|
| 833 |
+
"loss": 0.2013,
|
| 834 |
+
"step": 51500
|
| 835 |
+
},
|
| 836 |
+
{
|
| 837 |
+
"epoch": 2.7537997140284913,
|
| 838 |
+
"grad_norm": 4.8926496505737305,
|
| 839 |
+
"learning_rate": 0.00010833760313786463,
|
| 840 |
+
"loss": 0.2096,
|
| 841 |
+
"step": 52000
|
| 842 |
+
},
|
| 843 |
+
{
|
| 844 |
+
"epoch": 2.780278557432611,
|
| 845 |
+
"grad_norm": 1.4642279148101807,
|
| 846 |
+
"learning_rate": 0.00010745419530380396,
|
| 847 |
+
"loss": 0.2108,
|
| 848 |
+
"step": 52500
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 2.800296563046126,
|
| 852 |
+
"eval_loss": 0.2713735103607178,
|
| 853 |
+
"eval_runtime": 420.8891,
|
| 854 |
+
"eval_samples_per_second": 10.946,
|
| 855 |
+
"eval_steps_per_second": 5.474,
|
| 856 |
+
"step": 52878
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 2.8067574008367315,
|
| 860 |
+
"grad_norm": 2.818092107772827,
|
| 861 |
+
"learning_rate": 0.0001065707874697433,
|
| 862 |
+
"loss": 0.2011,
|
| 863 |
+
"step": 53000
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 2.8332362442408514,
|
| 867 |
+
"grad_norm": 2.3732500076293945,
|
| 868 |
+
"learning_rate": 0.00010568737963568261,
|
| 869 |
+
"loss": 0.2017,
|
| 870 |
+
"step": 53500
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 2.8597150876449717,
|
| 874 |
+
"grad_norm": 5.441200256347656,
|
| 875 |
+
"learning_rate": 0.00010480397180162194,
|
| 876 |
+
"loss": 0.2024,
|
| 877 |
+
"step": 54000
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 2.8861939310490916,
|
| 881 |
+
"grad_norm": 2.660677909851074,
|
| 882 |
+
"learning_rate": 0.00010392056396756127,
|
| 883 |
+
"loss": 0.2062,
|
| 884 |
+
"step": 54500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 2.912672774453212,
|
| 888 |
+
"grad_norm": 3.46864652633667,
|
| 889 |
+
"learning_rate": 0.00010303892294916872,
|
| 890 |
+
"loss": 0.1979,
|
| 891 |
+
"step": 55000
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 2.9391516178573323,
|
| 895 |
+
"grad_norm": 1.95028817653656,
|
| 896 |
+
"learning_rate": 0.00010215551511510804,
|
| 897 |
+
"loss": 0.1963,
|
| 898 |
+
"step": 55500
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 2.965630461261452,
|
| 902 |
+
"grad_norm": 2.8324038982391357,
|
| 903 |
+
"learning_rate": 0.00010127210728104737,
|
| 904 |
+
"loss": 0.2033,
|
| 905 |
+
"step": 56000
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 2.992109304665572,
|
| 909 |
+
"grad_norm": 2.3465640544891357,
|
| 910 |
+
"learning_rate": 0.0001003886994469867,
|
| 911 |
+
"loss": 0.2023,
|
| 912 |
+
"step": 56500
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 3.0003177461208494,
|
| 916 |
+
"eval_loss": 0.26497137546539307,
|
| 917 |
+
"eval_runtime": 421.6284,
|
| 918 |
+
"eval_samples_per_second": 10.927,
|
| 919 |
+
"eval_steps_per_second": 5.465,
|
| 920 |
+
"step": 56655
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 3.0185881480696923,
|
| 924 |
+
"grad_norm": 2.5295658111572266,
|
| 925 |
+
"learning_rate": 9.950705842859415e-05,
|
| 926 |
+
"loss": 0.1767,
|
| 927 |
+
"step": 57000
|
| 928 |
+
},
|
| 929 |
+
{
|
| 930 |
+
"epoch": 3.045066991473812,
|
| 931 |
+
"grad_norm": 2.448827028274536,
|
| 932 |
+
"learning_rate": 9.862365059453348e-05,
|
| 933 |
+
"loss": 0.1701,
|
| 934 |
+
"step": 57500
|
| 935 |
+
},
|
| 936 |
+
{
|
| 937 |
+
"epoch": 3.0715458348779325,
|
| 938 |
+
"grad_norm": 2.7891571521759033,
|
| 939 |
+
"learning_rate": 9.774200957614092e-05,
|
| 940 |
+
"loss": 0.1793,
|
| 941 |
+
"step": 58000
|
| 942 |
+
},
|
| 943 |
+
{
|
| 944 |
+
"epoch": 3.0980246782820524,
|
| 945 |
+
"grad_norm": 1.5115572214126587,
|
| 946 |
+
"learning_rate": 9.685860174208026e-05,
|
| 947 |
+
"loss": 0.1749,
|
| 948 |
+
"step": 58500
|
| 949 |
+
},
|
| 950 |
+
{
|
| 951 |
+
"epoch": 3.1245035216861727,
|
| 952 |
+
"grad_norm": 4.49403190612793,
|
| 953 |
+
"learning_rate": 9.597519390801958e-05,
|
| 954 |
+
"loss": 0.1666,
|
| 955 |
+
"step": 59000
|
| 956 |
+
},
|
| 957 |
+
{
|
| 958 |
+
"epoch": 3.150982365090293,
|
| 959 |
+
"grad_norm": 2.545595645904541,
|
| 960 |
+
"learning_rate": 9.509178607395891e-05,
|
| 961 |
+
"loss": 0.1755,
|
| 962 |
+
"step": 59500
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 3.177461208494413,
|
| 966 |
+
"grad_norm": 1.3802286386489868,
|
| 967 |
+
"learning_rate": 9.420837823989823e-05,
|
| 968 |
+
"loss": 0.1811,
|
| 969 |
+
"step": 60000
|
| 970 |
+
},
|
| 971 |
+
{
|
| 972 |
+
"epoch": 3.200338929195573,
|
| 973 |
+
"eval_loss": 0.2708809971809387,
|
| 974 |
+
"eval_runtime": 421.9193,
|
| 975 |
+
"eval_samples_per_second": 10.919,
|
| 976 |
+
"eval_steps_per_second": 5.461,
|
| 977 |
+
"step": 60432
|
| 978 |
+
},
|
| 979 |
+
{
|
| 980 |
+
"epoch": 3.2039400518985333,
|
| 981 |
+
"grad_norm": 0.7773789167404175,
|
| 982 |
+
"learning_rate": 9.332497040583756e-05,
|
| 983 |
+
"loss": 0.1714,
|
| 984 |
+
"step": 60500
|
| 985 |
+
},
|
| 986 |
+
{
|
| 987 |
+
"epoch": 3.230418895302653,
|
| 988 |
+
"grad_norm": 0.647718071937561,
|
| 989 |
+
"learning_rate": 9.244332938744501e-05,
|
| 990 |
+
"loss": 0.1742,
|
| 991 |
+
"step": 61000
|
| 992 |
+
},
|
| 993 |
+
{
|
| 994 |
+
"epoch": 3.2568977387067735,
|
| 995 |
+
"grad_norm": 2.515371322631836,
|
| 996 |
+
"learning_rate": 9.155992155338434e-05,
|
| 997 |
+
"loss": 0.1725,
|
| 998 |
+
"step": 61500
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 3.2833765821108933,
|
| 1002 |
+
"grad_norm": 3.261064291000366,
|
| 1003 |
+
"learning_rate": 9.067651371932367e-05,
|
| 1004 |
+
"loss": 0.1713,
|
| 1005 |
+
"step": 62000
|
| 1006 |
+
},
|
| 1007 |
+
{
|
| 1008 |
+
"epoch": 3.3098554255150137,
|
| 1009 |
+
"grad_norm": 2.5491161346435547,
|
| 1010 |
+
"learning_rate": 8.9793105885263e-05,
|
| 1011 |
+
"loss": 0.1702,
|
| 1012 |
+
"step": 62500
|
| 1013 |
+
},
|
| 1014 |
+
{
|
| 1015 |
+
"epoch": 3.3363342689191335,
|
| 1016 |
+
"grad_norm": 2.7108869552612305,
|
| 1017 |
+
"learning_rate": 8.890969805120232e-05,
|
| 1018 |
+
"loss": 0.1756,
|
| 1019 |
+
"step": 63000
|
| 1020 |
+
},
|
| 1021 |
+
{
|
| 1022 |
+
"epoch": 3.362813112323254,
|
| 1023 |
+
"grad_norm": 2.062349319458008,
|
| 1024 |
+
"learning_rate": 8.802629021714165e-05,
|
| 1025 |
+
"loss": 0.174,
|
| 1026 |
+
"step": 63500
|
| 1027 |
+
},
|
| 1028 |
+
{
|
| 1029 |
+
"epoch": 3.3892919557273737,
|
| 1030 |
+
"grad_norm": 0.7416903376579285,
|
| 1031 |
+
"learning_rate": 8.714288238308098e-05,
|
| 1032 |
+
"loss": 0.1702,
|
| 1033 |
+
"step": 64000
|
| 1034 |
+
},
|
| 1035 |
+
{
|
| 1036 |
+
"epoch": 3.400360112270296,
|
| 1037 |
+
"eval_loss": 0.26548436284065247,
|
| 1038 |
+
"eval_runtime": 421.3467,
|
| 1039 |
+
"eval_samples_per_second": 10.934,
|
| 1040 |
+
"eval_steps_per_second": 5.468,
|
| 1041 |
+
"step": 64209
|
| 1042 |
+
},
|
| 1043 |
+
{
|
| 1044 |
+
"epoch": 3.415770799131494,
|
| 1045 |
+
"grad_norm": 1.9311342239379883,
|
| 1046 |
+
"learning_rate": 8.625947454902031e-05,
|
| 1047 |
+
"loss": 0.1744,
|
| 1048 |
+
"step": 64500
|
| 1049 |
+
},
|
| 1050 |
+
{
|
| 1051 |
+
"epoch": 3.442249642535614,
|
| 1052 |
+
"grad_norm": 2.2730724811553955,
|
| 1053 |
+
"learning_rate": 8.537606671495964e-05,
|
| 1054 |
+
"loss": 0.172,
|
| 1055 |
+
"step": 65000
|
| 1056 |
+
},
|
| 1057 |
+
{
|
| 1058 |
+
"epoch": 3.4687284859397343,
|
| 1059 |
+
"grad_norm": 2.294510841369629,
|
| 1060 |
+
"learning_rate": 8.449442569656708e-05,
|
| 1061 |
+
"loss": 0.1732,
|
| 1062 |
+
"step": 65500
|
| 1063 |
+
},
|
| 1064 |
+
{
|
| 1065 |
+
"epoch": 3.495207329343854,
|
| 1066 |
+
"grad_norm": 2.0610592365264893,
|
| 1067 |
+
"learning_rate": 8.361101786250642e-05,
|
| 1068 |
+
"loss": 0.1761,
|
| 1069 |
+
"step": 66000
|
| 1070 |
+
},
|
| 1071 |
+
{
|
| 1072 |
+
"epoch": 3.5216861727479745,
|
| 1073 |
+
"grad_norm": 5.357800483703613,
|
| 1074 |
+
"learning_rate": 8.272761002844574e-05,
|
| 1075 |
+
"loss": 0.1773,
|
| 1076 |
+
"step": 66500
|
| 1077 |
+
},
|
| 1078 |
+
{
|
| 1079 |
+
"epoch": 3.5481650161520943,
|
| 1080 |
+
"grad_norm": 1.0783665180206299,
|
| 1081 |
+
"learning_rate": 8.184420219438507e-05,
|
| 1082 |
+
"loss": 0.1755,
|
| 1083 |
+
"step": 67000
|
| 1084 |
+
},
|
| 1085 |
+
{
|
| 1086 |
+
"epoch": 3.5746438595562147,
|
| 1087 |
+
"grad_norm": 2.4767355918884277,
|
| 1088 |
+
"learning_rate": 8.096079436032438e-05,
|
| 1089 |
+
"loss": 0.176,
|
| 1090 |
+
"step": 67500
|
| 1091 |
+
},
|
| 1092 |
+
{
|
| 1093 |
+
"epoch": 3.600381295345019,
|
| 1094 |
+
"eval_loss": 0.26650214195251465,
|
| 1095 |
+
"eval_runtime": 421.0598,
|
| 1096 |
+
"eval_samples_per_second": 10.941,
|
| 1097 |
+
"eval_steps_per_second": 5.472,
|
| 1098 |
+
"step": 67986
|
| 1099 |
+
},
|
| 1100 |
+
{
|
| 1101 |
+
"epoch": 3.6011227029603345,
|
| 1102 |
+
"grad_norm": 3.160248041152954,
|
| 1103 |
+
"learning_rate": 8.007915334193185e-05,
|
| 1104 |
+
"loss": 0.1671,
|
| 1105 |
+
"step": 68000
|
| 1106 |
+
},
|
| 1107 |
+
{
|
| 1108 |
+
"epoch": 3.627601546364455,
|
| 1109 |
+
"grad_norm": 2.2208316326141357,
|
| 1110 |
+
"learning_rate": 7.919574550787116e-05,
|
| 1111 |
+
"loss": 0.1766,
|
| 1112 |
+
"step": 68500
|
| 1113 |
+
},
|
| 1114 |
+
{
|
| 1115 |
+
"epoch": 3.6540803897685747,
|
| 1116 |
+
"grad_norm": 3.152665615081787,
|
| 1117 |
+
"learning_rate": 7.83123376738105e-05,
|
| 1118 |
+
"loss": 0.18,
|
| 1119 |
+
"step": 69000
|
| 1120 |
+
},
|
| 1121 |
+
{
|
| 1122 |
+
"epoch": 3.680559233172695,
|
| 1123 |
+
"grad_norm": 0.9788137078285217,
|
| 1124 |
+
"learning_rate": 7.742892983974983e-05,
|
| 1125 |
+
"loss": 0.1721,
|
| 1126 |
+
"step": 69500
|
| 1127 |
+
},
|
| 1128 |
+
{
|
| 1129 |
+
"epoch": 3.707038076576815,
|
| 1130 |
+
"grad_norm": 1.457961082458496,
|
| 1131 |
+
"learning_rate": 7.654728882135728e-05,
|
| 1132 |
+
"loss": 0.1694,
|
| 1133 |
+
"step": 70000
|
| 1134 |
+
},
|
| 1135 |
+
{
|
| 1136 |
+
"epoch": 3.7335169199809353,
|
| 1137 |
+
"grad_norm": 1.4032424688339233,
|
| 1138 |
+
"learning_rate": 7.56638809872966e-05,
|
| 1139 |
+
"loss": 0.1633,
|
| 1140 |
+
"step": 70500
|
| 1141 |
+
},
|
| 1142 |
+
{
|
| 1143 |
+
"epoch": 3.7599957633850556,
|
| 1144 |
+
"grad_norm": 4.617145538330078,
|
| 1145 |
+
"learning_rate": 7.478223996890404e-05,
|
| 1146 |
+
"loss": 0.1785,
|
| 1147 |
+
"step": 71000
|
| 1148 |
+
},
|
| 1149 |
+
{
|
| 1150 |
+
"epoch": 3.7864746067891755,
|
| 1151 |
+
"grad_norm": 1.1956781148910522,
|
| 1152 |
+
"learning_rate": 7.389883213484337e-05,
|
| 1153 |
+
"loss": 0.1702,
|
| 1154 |
+
"step": 71500
|
| 1155 |
+
},
|
| 1156 |
+
{
|
| 1157 |
+
"epoch": 3.8004024784197425,
|
| 1158 |
+
"eval_loss": 0.26657551527023315,
|
| 1159 |
+
"eval_runtime": 422.2898,
|
| 1160 |
+
"eval_samples_per_second": 10.91,
|
| 1161 |
+
"eval_steps_per_second": 5.456,
|
| 1162 |
+
"step": 71763
|
| 1163 |
+
},
|
| 1164 |
+
{
|
| 1165 |
+
"epoch": 3.8129534501932953,
|
| 1166 |
+
"grad_norm": 0.6750999093055725,
|
| 1167 |
+
"learning_rate": 7.301542430078269e-05,
|
| 1168 |
+
"loss": 0.1775,
|
| 1169 |
+
"step": 72000
|
| 1170 |
+
},
|
| 1171 |
+
{
|
| 1172 |
+
"epoch": 3.8394322935974157,
|
| 1173 |
+
"grad_norm": 1.8476779460906982,
|
| 1174 |
+
"learning_rate": 7.213201646672204e-05,
|
| 1175 |
+
"loss": 0.1675,
|
| 1176 |
+
"step": 72500
|
| 1177 |
+
},
|
| 1178 |
+
{
|
| 1179 |
+
"epoch": 3.865911137001536,
|
| 1180 |
+
"grad_norm": 2.567692756652832,
|
| 1181 |
+
"learning_rate": 7.124860863266137e-05,
|
| 1182 |
+
"loss": 0.1723,
|
| 1183 |
+
"step": 73000
|
| 1184 |
+
},
|
| 1185 |
+
{
|
| 1186 |
+
"epoch": 3.892389980405656,
|
| 1187 |
+
"grad_norm": 1.9294815063476562,
|
| 1188 |
+
"learning_rate": 7.036520079860068e-05,
|
| 1189 |
+
"loss": 0.1703,
|
| 1190 |
+
"step": 73500
|
| 1191 |
+
},
|
| 1192 |
+
{
|
| 1193 |
+
"epoch": 3.9188688238097757,
|
| 1194 |
+
"grad_norm": 2.111727714538574,
|
| 1195 |
+
"learning_rate": 6.948355978020813e-05,
|
| 1196 |
+
"loss": 0.1686,
|
| 1197 |
+
"step": 74000
|
| 1198 |
+
},
|
| 1199 |
+
{
|
| 1200 |
+
"epoch": 3.945347667213896,
|
| 1201 |
+
"grad_norm": 2.216778039932251,
|
| 1202 |
+
"learning_rate": 6.860015194614746e-05,
|
| 1203 |
+
"loss": 0.1725,
|
| 1204 |
+
"step": 74500
|
| 1205 |
+
},
|
| 1206 |
+
{
|
| 1207 |
+
"epoch": 3.9718265106180164,
|
| 1208 |
+
"grad_norm": 1.494307041168213,
|
| 1209 |
+
"learning_rate": 6.77167441120868e-05,
|
| 1210 |
+
"loss": 0.1789,
|
| 1211 |
+
"step": 75000
|
| 1212 |
+
},
|
| 1213 |
+
{
|
| 1214 |
+
"epoch": 3.9983053540221363,
|
| 1215 |
+
"grad_norm": 1.8370423316955566,
|
| 1216 |
+
"learning_rate": 6.683333627802611e-05,
|
| 1217 |
+
"loss": 0.1597,
|
| 1218 |
+
"step": 75500
|
| 1219 |
+
},
|
| 1220 |
+
{
|
| 1221 |
+
"epoch": 4.000423661494466,
|
| 1222 |
+
"eval_loss": 0.26198258996009827,
|
| 1223 |
+
"eval_runtime": 422.2011,
|
| 1224 |
+
"eval_samples_per_second": 10.912,
|
| 1225 |
+
"eval_steps_per_second": 5.457,
|
| 1226 |
+
"step": 75540
|
| 1227 |
+
},
|
| 1228 |
+
{
|
| 1229 |
+
"epoch": 4.024784197426256,
|
| 1230 |
+
"grad_norm": 2.989509105682373,
|
| 1231 |
+
"learning_rate": 6.595169525963356e-05,
|
| 1232 |
+
"loss": 0.1527,
|
| 1233 |
+
"step": 76000
|
| 1234 |
+
},
|
| 1235 |
+
{
|
| 1236 |
+
"epoch": 4.051263040830377,
|
| 1237 |
+
"grad_norm": 2.3569653034210205,
|
| 1238 |
+
"learning_rate": 6.506828742557289e-05,
|
| 1239 |
+
"loss": 0.1481,
|
| 1240 |
+
"step": 76500
|
| 1241 |
+
},
|
| 1242 |
+
{
|
| 1243 |
+
"epoch": 4.077741884234497,
|
| 1244 |
+
"grad_norm": 1.7681636810302734,
|
| 1245 |
+
"learning_rate": 6.418487959151222e-05,
|
| 1246 |
+
"loss": 0.1438,
|
| 1247 |
+
"step": 77000
|
| 1248 |
+
},
|
| 1249 |
+
{
|
| 1250 |
+
"epoch": 4.104220727638617,
|
| 1251 |
+
"grad_norm": 1.9401808977127075,
|
| 1252 |
+
"learning_rate": 6.330147175745155e-05,
|
| 1253 |
+
"loss": 0.1453,
|
| 1254 |
+
"step": 77500
|
| 1255 |
+
},
|
| 1256 |
+
{
|
| 1257 |
+
"epoch": 4.1306995710427366,
|
| 1258 |
+
"grad_norm": 1.9549821615219116,
|
| 1259 |
+
"learning_rate": 6.241806392339088e-05,
|
| 1260 |
+
"loss": 0.1453,
|
| 1261 |
+
"step": 78000
|
| 1262 |
+
},
|
| 1263 |
+
{
|
| 1264 |
+
"epoch": 4.157178414446857,
|
| 1265 |
+
"grad_norm": 2.7086589336395264,
|
| 1266 |
+
"learning_rate": 6.15346560893302e-05,
|
| 1267 |
+
"loss": 0.1424,
|
| 1268 |
+
"step": 78500
|
| 1269 |
+
},
|
| 1270 |
+
{
|
| 1271 |
+
"epoch": 4.183657257850977,
|
| 1272 |
+
"grad_norm": 2.071589469909668,
|
| 1273 |
+
"learning_rate": 6.065124825526953e-05,
|
| 1274 |
+
"loss": 0.1388,
|
| 1275 |
+
"step": 79000
|
| 1276 |
+
},
|
| 1277 |
+
{
|
| 1278 |
+
"epoch": 4.200444844569189,
|
| 1279 |
+
"eval_loss": 0.2703910171985626,
|
| 1280 |
+
"eval_runtime": 422.189,
|
| 1281 |
+
"eval_samples_per_second": 10.912,
|
| 1282 |
+
"eval_steps_per_second": 5.457,
|
| 1283 |
+
"step": 79317
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 4.210136101255097,
|
| 1287 |
+
"grad_norm": 2.643260955810547,
|
| 1288 |
+
"learning_rate": 5.9767840421208854e-05,
|
| 1289 |
+
"loss": 0.1501,
|
| 1290 |
+
"step": 79500
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 4.236614944659217,
|
| 1294 |
+
"grad_norm": 1.3025317192077637,
|
| 1295 |
+
"learning_rate": 5.888796621848443e-05,
|
| 1296 |
+
"loss": 0.1531,
|
| 1297 |
+
"step": 80000
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 4.263093788063338,
|
| 1301 |
+
"grad_norm": 0.5772294998168945,
|
| 1302 |
+
"learning_rate": 5.800455838442376e-05,
|
| 1303 |
+
"loss": 0.1547,
|
| 1304 |
+
"step": 80500
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 4.289572631467458,
|
| 1308 |
+
"grad_norm": 1.2284057140350342,
|
| 1309 |
+
"learning_rate": 5.712115055036308e-05,
|
| 1310 |
+
"loss": 0.1532,
|
| 1311 |
+
"step": 81000
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 4.3160514748715775,
|
| 1315 |
+
"grad_norm": 1.778266429901123,
|
| 1316 |
+
"learning_rate": 5.623774271630241e-05,
|
| 1317 |
+
"loss": 0.1444,
|
| 1318 |
+
"step": 81500
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 4.342530318275697,
|
| 1322 |
+
"grad_norm": 4.491095542907715,
|
| 1323 |
+
"learning_rate": 5.535433488224174e-05,
|
| 1324 |
+
"loss": 0.141,
|
| 1325 |
+
"step": 82000
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 4.369009161679818,
|
| 1329 |
+
"grad_norm": 1.7908458709716797,
|
| 1330 |
+
"learning_rate": 5.4470927048181065e-05,
|
| 1331 |
+
"loss": 0.1521,
|
| 1332 |
+
"step": 82500
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 4.395488005083938,
|
| 1336 |
+
"grad_norm": 0.9978257417678833,
|
| 1337 |
+
"learning_rate": 5.3587519214120395e-05,
|
| 1338 |
+
"loss": 0.1452,
|
| 1339 |
+
"step": 83000
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 4.400466027643913,
|
| 1343 |
+
"eval_loss": 0.27073460817337036,
|
| 1344 |
+
"eval_runtime": 422.1641,
|
| 1345 |
+
"eval_samples_per_second": 10.913,
|
| 1346 |
+
"eval_steps_per_second": 5.458,
|
| 1347 |
+
"step": 83094
|
| 1348 |
+
},
|
| 1349 |
+
{
|
| 1350 |
+
"epoch": 4.421966848488058,
|
| 1351 |
+
"grad_norm": 3.2376346588134766,
|
| 1352 |
+
"learning_rate": 5.270411138005972e-05,
|
| 1353 |
+
"loss": 0.1487,
|
| 1354 |
+
"step": 83500
|
| 1355 |
+
},
|
| 1356 |
+
{
|
| 1357 |
+
"epoch": 4.448445691892178,
|
| 1358 |
+
"grad_norm": 1.5663423538208008,
|
| 1359 |
+
"learning_rate": 5.182070354599905e-05,
|
| 1360 |
+
"loss": 0.1489,
|
| 1361 |
+
"step": 84000
|
| 1362 |
+
},
|
| 1363 |
+
{
|
| 1364 |
+
"epoch": 4.4749245352962985,
|
| 1365 |
+
"grad_norm": 1.9214493036270142,
|
| 1366 |
+
"learning_rate": 5.093729571193837e-05,
|
| 1367 |
+
"loss": 0.1431,
|
| 1368 |
+
"step": 84500
|
| 1369 |
+
},
|
| 1370 |
+
{
|
| 1371 |
+
"epoch": 4.501403378700418,
|
| 1372 |
+
"grad_norm": 0.39390167593955994,
|
| 1373 |
+
"learning_rate": 5.005565469354583e-05,
|
| 1374 |
+
"loss": 0.1488,
|
| 1375 |
+
"step": 85000
|
| 1376 |
+
},
|
| 1377 |
+
{
|
| 1378 |
+
"epoch": 4.527882222104538,
|
| 1379 |
+
"grad_norm": 1.417643427848816,
|
| 1380 |
+
"learning_rate": 4.9172246859485154e-05,
|
| 1381 |
+
"loss": 0.1424,
|
| 1382 |
+
"step": 85500
|
| 1383 |
+
},
|
| 1384 |
+
{
|
| 1385 |
+
"epoch": 4.554361065508658,
|
| 1386 |
+
"grad_norm": 1.4273784160614014,
|
| 1387 |
+
"learning_rate": 4.8288839025424484e-05,
|
| 1388 |
+
"loss": 0.1426,
|
| 1389 |
+
"step": 86000
|
| 1390 |
+
},
|
| 1391 |
+
{
|
| 1392 |
+
"epoch": 4.580839908912779,
|
| 1393 |
+
"grad_norm": 2.3503715991973877,
|
| 1394 |
+
"learning_rate": 4.740543119136381e-05,
|
| 1395 |
+
"loss": 0.146,
|
| 1396 |
+
"step": 86500
|
| 1397 |
+
},
|
| 1398 |
+
{
|
| 1399 |
+
"epoch": 4.600487210718636,
|
| 1400 |
+
"eval_loss": 0.2704991102218628,
|
| 1401 |
+
"eval_runtime": 422.5207,
|
| 1402 |
+
"eval_samples_per_second": 10.904,
|
| 1403 |
+
"eval_steps_per_second": 5.453,
|
| 1404 |
+
"step": 86871
|
| 1405 |
+
},
|
| 1406 |
+
{
|
| 1407 |
+
"epoch": 4.607318752316899,
|
| 1408 |
+
"grad_norm": 2.605318784713745,
|
| 1409 |
+
"learning_rate": 4.652202335730314e-05,
|
| 1410 |
+
"loss": 0.1496,
|
| 1411 |
+
"step": 87000
|
| 1412 |
+
},
|
| 1413 |
+
{
|
| 1414 |
+
"epoch": 4.633797595721019,
|
| 1415 |
+
"grad_norm": 2.3824856281280518,
|
| 1416 |
+
"learning_rate": 4.563861552324246e-05,
|
| 1417 |
+
"loss": 0.1389,
|
| 1418 |
+
"step": 87500
|
| 1419 |
+
},
|
| 1420 |
+
{
|
| 1421 |
+
"epoch": 4.6602764391251394,
|
| 1422 |
+
"grad_norm": 1.4241347312927246,
|
| 1423 |
+
"learning_rate": 4.4755207689181785e-05,
|
| 1424 |
+
"loss": 0.1451,
|
| 1425 |
+
"step": 88000
|
| 1426 |
+
},
|
| 1427 |
+
{
|
| 1428 |
+
"epoch": 4.686755282529259,
|
| 1429 |
+
"grad_norm": 1.1952823400497437,
|
| 1430 |
+
"learning_rate": 4.3871799855121115e-05,
|
| 1431 |
+
"loss": 0.1441,
|
| 1432 |
+
"step": 88500
|
| 1433 |
+
},
|
| 1434 |
+
{
|
| 1435 |
+
"epoch": 4.713234125933379,
|
| 1436 |
+
"grad_norm": 1.2604337930679321,
|
| 1437 |
+
"learning_rate": 4.299015883672857e-05,
|
| 1438 |
+
"loss": 0.144,
|
| 1439 |
+
"step": 89000
|
| 1440 |
+
},
|
| 1441 |
+
{
|
| 1442 |
+
"epoch": 4.739712969337499,
|
| 1443 |
+
"grad_norm": 2.8621230125427246,
|
| 1444 |
+
"learning_rate": 4.210675100266789e-05,
|
| 1445 |
+
"loss": 0.1481,
|
| 1446 |
+
"step": 89500
|
| 1447 |
+
},
|
| 1448 |
+
{
|
| 1449 |
+
"epoch": 4.766191812741619,
|
| 1450 |
+
"grad_norm": 0.4195690453052521,
|
| 1451 |
+
"learning_rate": 4.122334316860722e-05,
|
| 1452 |
+
"loss": 0.1505,
|
| 1453 |
+
"step": 90000
|
| 1454 |
+
},
|
| 1455 |
+
{
|
| 1456 |
+
"epoch": 4.79267065614574,
|
| 1457 |
+
"grad_norm": 2.966862916946411,
|
| 1458 |
+
"learning_rate": 4.0339935334546544e-05,
|
| 1459 |
+
"loss": 0.1388,
|
| 1460 |
+
"step": 90500
|
| 1461 |
+
},
|
| 1462 |
+
{
|
| 1463 |
+
"epoch": 4.8005083937933595,
|
| 1464 |
+
"eval_loss": 0.26222139596939087,
|
| 1465 |
+
"eval_runtime": 421.6806,
|
| 1466 |
+
"eval_samples_per_second": 10.925,
|
| 1467 |
+
"eval_steps_per_second": 5.464,
|
| 1468 |
+
"step": 90648
|
| 1469 |
+
},
|
| 1470 |
+
{
|
| 1471 |
+
"epoch": 4.81914949954986,
|
| 1472 |
+
"grad_norm": 1.312051773071289,
|
| 1473 |
+
"learning_rate": 3.9458294316153995e-05,
|
| 1474 |
+
"loss": 0.1428,
|
| 1475 |
+
"step": 91000
|
| 1476 |
+
},
|
| 1477 |
+
{
|
| 1478 |
+
"epoch": 4.8456283429539795,
|
| 1479 |
+
"grad_norm": 0.6421635746955872,
|
| 1480 |
+
"learning_rate": 3.8574886482093326e-05,
|
| 1481 |
+
"loss": 0.1435,
|
| 1482 |
+
"step": 91500
|
| 1483 |
+
},
|
| 1484 |
+
{
|
| 1485 |
+
"epoch": 4.8721071863581,
|
| 1486 |
+
"grad_norm": 1.8707529306411743,
|
| 1487 |
+
"learning_rate": 3.769147864803265e-05,
|
| 1488 |
+
"loss": 0.1423,
|
| 1489 |
+
"step": 92000
|
| 1490 |
+
},
|
| 1491 |
+
{
|
| 1492 |
+
"epoch": 4.89858602976222,
|
| 1493 |
+
"grad_norm": 1.6387176513671875,
|
| 1494 |
+
"learning_rate": 3.680807081397198e-05,
|
| 1495 |
+
"loss": 0.1378,
|
| 1496 |
+
"step": 92500
|
| 1497 |
+
},
|
| 1498 |
+
{
|
| 1499 |
+
"epoch": 4.92506487316634,
|
| 1500 |
+
"grad_norm": 2.160985231399536,
|
| 1501 |
+
"learning_rate": 3.592642979557943e-05,
|
| 1502 |
+
"loss": 0.1439,
|
| 1503 |
+
"step": 93000
|
| 1504 |
+
},
|
| 1505 |
+
{
|
| 1506 |
+
"epoch": 4.95154371657046,
|
| 1507 |
+
"grad_norm": 1.3969603776931763,
|
| 1508 |
+
"learning_rate": 3.5043021961518754e-05,
|
| 1509 |
+
"loss": 0.1436,
|
| 1510 |
+
"step": 93500
|
| 1511 |
+
},
|
| 1512 |
+
{
|
| 1513 |
+
"epoch": 4.978022559974581,
|
| 1514 |
+
"grad_norm": 1.672614574432373,
|
| 1515 |
+
"learning_rate": 3.4159614127458085e-05,
|
| 1516 |
+
"loss": 0.1435,
|
| 1517 |
+
"step": 94000
|
| 1518 |
+
},
|
| 1519 |
+
{
|
| 1520 |
+
"epoch": 5.000529576868082,
|
| 1521 |
+
"eval_loss": 0.26488494873046875,
|
| 1522 |
+
"eval_runtime": 422.1282,
|
| 1523 |
+
"eval_samples_per_second": 10.914,
|
| 1524 |
+
"eval_steps_per_second": 5.458,
|
| 1525 |
+
"step": 94425
|
| 1526 |
+
},
|
| 1527 |
+
{
|
| 1528 |
+
"epoch": 5.0045014033787005,
|
| 1529 |
+
"grad_norm": 3.6483752727508545,
|
| 1530 |
+
"learning_rate": 3.327620629339741e-05,
|
| 1531 |
+
"loss": 0.1358,
|
| 1532 |
+
"step": 94500
|
| 1533 |
+
},
|
| 1534 |
+
{
|
| 1535 |
+
"epoch": 5.03098024678282,
|
| 1536 |
+
"grad_norm": 1.201677918434143,
|
| 1537 |
+
"learning_rate": 3.239279845933674e-05,
|
| 1538 |
+
"loss": 0.1302,
|
| 1539 |
+
"step": 95000
|
| 1540 |
+
},
|
| 1541 |
+
{
|
| 1542 |
+
"epoch": 5.05745909018694,
|
| 1543 |
+
"grad_norm": 2.516324281692505,
|
| 1544 |
+
"learning_rate": 3.150939062527606e-05,
|
| 1545 |
+
"loss": 0.1218,
|
| 1546 |
+
"step": 95500
|
| 1547 |
+
},
|
| 1548 |
+
{
|
| 1549 |
+
"epoch": 5.083937933591061,
|
| 1550 |
+
"grad_norm": 0.14960531890392303,
|
| 1551 |
+
"learning_rate": 3.06259827912154e-05,
|
| 1552 |
+
"loss": 0.1228,
|
| 1553 |
+
"step": 96000
|
| 1554 |
+
},
|
| 1555 |
+
{
|
| 1556 |
+
"epoch": 5.110416776995181,
|
| 1557 |
+
"grad_norm": 0.23572592437267303,
|
| 1558 |
+
"learning_rate": 2.9742574957154723e-05,
|
| 1559 |
+
"loss": 0.1302,
|
| 1560 |
+
"step": 96500
|
| 1561 |
+
},
|
| 1562 |
+
{
|
| 1563 |
+
"epoch": 5.136895620399301,
|
| 1564 |
+
"grad_norm": 1.1999266147613525,
|
| 1565 |
+
"learning_rate": 2.885916712309405e-05,
|
| 1566 |
+
"loss": 0.1201,
|
| 1567 |
+
"step": 97000
|
| 1568 |
+
},
|
| 1569 |
+
{
|
| 1570 |
+
"epoch": 5.163374463803421,
|
| 1571 |
+
"grad_norm": 2.529439687728882,
|
| 1572 |
+
"learning_rate": 2.7977526104701497e-05,
|
| 1573 |
+
"loss": 0.1217,
|
| 1574 |
+
"step": 97500
|
| 1575 |
+
},
|
| 1576 |
+
{
|
| 1577 |
+
"epoch": 5.1898533072075415,
|
| 1578 |
+
"grad_norm": 4.092609405517578,
|
| 1579 |
+
"learning_rate": 2.7094118270640828e-05,
|
| 1580 |
+
"loss": 0.1295,
|
| 1581 |
+
"step": 98000
|
| 1582 |
+
},
|
| 1583 |
+
{
|
| 1584 |
+
"epoch": 5.200550759942805,
|
| 1585 |
+
"eval_loss": 0.2793600857257843,
|
| 1586 |
+
"eval_runtime": 422.1911,
|
| 1587 |
+
"eval_samples_per_second": 10.912,
|
| 1588 |
+
"eval_steps_per_second": 5.457,
|
| 1589 |
+
"step": 98202
|
| 1590 |
+
},
|
| 1591 |
+
{
|
| 1592 |
+
"epoch": 5.216332150611661,
|
| 1593 |
+
"grad_norm": 1.677905797958374,
|
| 1594 |
+
"learning_rate": 2.6210710436580155e-05,
|
| 1595 |
+
"loss": 0.1232,
|
| 1596 |
+
"step": 98500
|
| 1597 |
+
},
|
| 1598 |
+
{
|
| 1599 |
+
"epoch": 5.242810994015781,
|
| 1600 |
+
"grad_norm": 2.0790467262268066,
|
| 1601 |
+
"learning_rate": 2.532730260251948e-05,
|
| 1602 |
+
"loss": 0.1145,
|
| 1603 |
+
"step": 99000
|
| 1604 |
+
},
|
| 1605 |
+
{
|
| 1606 |
+
"epoch": 5.269289837419901,
|
| 1607 |
+
"grad_norm": 0.821835458278656,
|
| 1608 |
+
"learning_rate": 2.444389476845881e-05,
|
| 1609 |
+
"loss": 0.122,
|
| 1610 |
+
"step": 99500
|
| 1611 |
+
},
|
| 1612 |
+
{
|
| 1613 |
+
"epoch": 5.295768680824022,
|
| 1614 |
+
"grad_norm": 1.6524577140808105,
|
| 1615 |
+
"learning_rate": 2.3560486934398135e-05,
|
| 1616 |
+
"loss": 0.1211,
|
| 1617 |
+
"step": 100000
|
| 1618 |
+
},
|
| 1619 |
+
{
|
| 1620 |
+
"epoch": 5.322247524228142,
|
| 1621 |
+
"grad_norm": 1.0877349376678467,
|
| 1622 |
+
"learning_rate": 2.2677079100337462e-05,
|
| 1623 |
+
"loss": 0.1213,
|
| 1624 |
+
"step": 100500
|
| 1625 |
+
},
|
| 1626 |
+
{
|
| 1627 |
+
"epoch": 5.348726367632262,
|
| 1628 |
+
"grad_norm": 1.8999888896942139,
|
| 1629 |
+
"learning_rate": 2.1793671266276793e-05,
|
| 1630 |
+
"loss": 0.1212,
|
| 1631 |
+
"step": 101000
|
| 1632 |
+
},
|
| 1633 |
+
{
|
| 1634 |
+
"epoch": 5.375205211036382,
|
| 1635 |
+
"grad_norm": 1.6684705018997192,
|
| 1636 |
+
"learning_rate": 2.091203024788424e-05,
|
| 1637 |
+
"loss": 0.1195,
|
| 1638 |
+
"step": 101500
|
| 1639 |
+
},
|
| 1640 |
+
{
|
| 1641 |
+
"epoch": 5.400571943017529,
|
| 1642 |
+
"eval_loss": 0.2780030369758606,
|
| 1643 |
+
"eval_runtime": 423.5819,
|
| 1644 |
+
"eval_samples_per_second": 10.876,
|
| 1645 |
+
"eval_steps_per_second": 5.439,
|
| 1646 |
+
"step": 101979
|
| 1647 |
+
},
|
| 1648 |
+
{
|
| 1649 |
+
"epoch": 5.401684054440502,
|
| 1650 |
+
"grad_norm": 4.362273216247559,
|
| 1651 |
+
"learning_rate": 2.0028622413823567e-05,
|
| 1652 |
+
"loss": 0.1269,
|
| 1653 |
+
"step": 102000
|
| 1654 |
+
},
|
| 1655 |
+
{
|
| 1656 |
+
"epoch": 5.428162897844622,
|
| 1657 |
+
"grad_norm": 1.8318756818771362,
|
| 1658 |
+
"learning_rate": 1.9146981395431015e-05,
|
| 1659 |
+
"loss": 0.1211,
|
| 1660 |
+
"step": 102500
|
| 1661 |
+
},
|
| 1662 |
+
{
|
| 1663 |
+
"epoch": 5.454641741248742,
|
| 1664 |
+
"grad_norm": 0.10318777710199356,
|
| 1665 |
+
"learning_rate": 1.8263573561370342e-05,
|
| 1666 |
+
"loss": 0.122,
|
| 1667 |
+
"step": 103000
|
| 1668 |
+
},
|
| 1669 |
+
{
|
| 1670 |
+
"epoch": 5.481120584652862,
|
| 1671 |
+
"grad_norm": 3.140679121017456,
|
| 1672 |
+
"learning_rate": 1.7380165727309673e-05,
|
| 1673 |
+
"loss": 0.1235,
|
| 1674 |
+
"step": 103500
|
| 1675 |
+
},
|
| 1676 |
+
{
|
| 1677 |
+
"epoch": 5.507599428056983,
|
| 1678 |
+
"grad_norm": 1.0891458988189697,
|
| 1679 |
+
"learning_rate": 1.6496757893249e-05,
|
| 1680 |
+
"loss": 0.1195,
|
| 1681 |
+
"step": 104000
|
| 1682 |
+
},
|
| 1683 |
+
{
|
| 1684 |
+
"epoch": 5.5340782714611025,
|
| 1685 |
+
"grad_norm": 1.6873986721038818,
|
| 1686 |
+
"learning_rate": 1.5613350059188326e-05,
|
| 1687 |
+
"loss": 0.1253,
|
| 1688 |
+
"step": 104500
|
| 1689 |
+
},
|
| 1690 |
+
{
|
| 1691 |
+
"epoch": 5.560557114865222,
|
| 1692 |
+
"grad_norm": 0.6695570945739746,
|
| 1693 |
+
"learning_rate": 1.4729942225127652e-05,
|
| 1694 |
+
"loss": 0.1245,
|
| 1695 |
+
"step": 105000
|
| 1696 |
+
},
|
| 1697 |
+
{
|
| 1698 |
+
"epoch": 5.587035958269343,
|
| 1699 |
+
"grad_norm": 1.3430531024932861,
|
| 1700 |
+
"learning_rate": 1.3846534391066982e-05,
|
| 1701 |
+
"loss": 0.124,
|
| 1702 |
+
"step": 105500
|
| 1703 |
+
},
|
| 1704 |
+
{
|
| 1705 |
+
"epoch": 5.600593126092252,
|
| 1706 |
+
"eval_loss": 0.27956077456474304,
|
| 1707 |
+
"eval_runtime": 422.629,
|
| 1708 |
+
"eval_samples_per_second": 10.901,
|
| 1709 |
+
"eval_steps_per_second": 5.452,
|
| 1710 |
+
"step": 105756
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 5.613514801673463,
|
| 1714 |
+
"grad_norm": 3.5196123123168945,
|
| 1715 |
+
"learning_rate": 1.2963126557006309e-05,
|
| 1716 |
+
"loss": 0.1259,
|
| 1717 |
+
"step": 106000
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 5.639993645077583,
|
| 1721 |
+
"grad_norm": 1.313728928565979,
|
| 1722 |
+
"learning_rate": 1.2081485538613757e-05,
|
| 1723 |
+
"loss": 0.124,
|
| 1724 |
+
"step": 106500
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 5.666472488481703,
|
| 1728 |
+
"grad_norm": 1.7873483896255493,
|
| 1729 |
+
"learning_rate": 1.1198077704553085e-05,
|
| 1730 |
+
"loss": 0.1297,
|
| 1731 |
+
"step": 107000
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 5.692951331885824,
|
| 1735 |
+
"grad_norm": 3.823012351989746,
|
| 1736 |
+
"learning_rate": 1.0314669870492412e-05,
|
| 1737 |
+
"loss": 0.1192,
|
| 1738 |
+
"step": 107500
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 5.7194301752899435,
|
| 1742 |
+
"grad_norm": 0.8463395833969116,
|
| 1743 |
+
"learning_rate": 9.43126203643174e-06,
|
| 1744 |
+
"loss": 0.1183,
|
| 1745 |
+
"step": 108000
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 5.745909018694063,
|
| 1749 |
+
"grad_norm": 3.8344569206237793,
|
| 1750 |
+
"learning_rate": 8.547854202371066e-06,
|
| 1751 |
+
"loss": 0.1207,
|
| 1752 |
+
"step": 108500
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 5.772387862098183,
|
| 1756 |
+
"grad_norm": 2.5786736011505127,
|
| 1757 |
+
"learning_rate": 7.666213183978516e-06,
|
| 1758 |
+
"loss": 0.1148,
|
| 1759 |
+
"step": 109000
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 5.798866705502304,
|
| 1763 |
+
"grad_norm": 2.022777795791626,
|
| 1764 |
+
"learning_rate": 6.7828053499178435e-06,
|
| 1765 |
+
"loss": 0.1183,
|
| 1766 |
+
"step": 109500
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 5.800614309166976,
|
| 1770 |
+
"eval_loss": 0.283890038728714,
|
| 1771 |
+
"eval_runtime": 421.7577,
|
| 1772 |
+
"eval_samples_per_second": 10.923,
|
| 1773 |
+
"eval_steps_per_second": 5.463,
|
| 1774 |
+
"step": 109533
|
| 1775 |
+
},
|
| 1776 |
+
{
|
| 1777 |
+
"epoch": 5.825345548906424,
|
| 1778 |
+
"grad_norm": 2.0010786056518555,
|
| 1779 |
+
"learning_rate": 5.9011643315252915e-06,
|
| 1780 |
+
"loss": 0.1184,
|
| 1781 |
+
"step": 110000
|
| 1782 |
+
},
|
| 1783 |
+
{
|
| 1784 |
+
"epoch": 5.851824392310544,
|
| 1785 |
+
"grad_norm": 3.0789356231689453,
|
| 1786 |
+
"learning_rate": 5.019523313132741e-06,
|
| 1787 |
+
"loss": 0.1231,
|
| 1788 |
+
"step": 110500
|
| 1789 |
+
},
|
| 1790 |
+
{
|
| 1791 |
+
"epoch": 5.878303235714664,
|
| 1792 |
+
"grad_norm": 0.8194056153297424,
|
| 1793 |
+
"learning_rate": 4.136115479072069e-06,
|
| 1794 |
+
"loss": 0.119,
|
| 1795 |
+
"step": 111000
|
| 1796 |
+
},
|
| 1797 |
+
{
|
| 1798 |
+
"epoch": 5.904782079118784,
|
| 1799 |
+
"grad_norm": 1.4833017587661743,
|
| 1800 |
+
"learning_rate": 3.2527076450113963e-06,
|
| 1801 |
+
"loss": 0.119,
|
| 1802 |
+
"step": 111500
|
| 1803 |
+
},
|
| 1804 |
+
{
|
| 1805 |
+
"epoch": 5.931260922522904,
|
| 1806 |
+
"grad_norm": 3.702115535736084,
|
| 1807 |
+
"learning_rate": 2.3692998109507236e-06,
|
| 1808 |
+
"loss": 0.1129,
|
| 1809 |
+
"step": 112000
|
| 1810 |
+
},
|
| 1811 |
+
{
|
| 1812 |
+
"epoch": 5.957739765927024,
|
| 1813 |
+
"grad_norm": 2.612248659133911,
|
| 1814 |
+
"learning_rate": 1.4858919768900512e-06,
|
| 1815 |
+
"loss": 0.1106,
|
| 1816 |
+
"step": 112500
|
| 1817 |
+
},
|
| 1818 |
+
{
|
| 1819 |
+
"epoch": 5.984218609331144,
|
| 1820 |
+
"grad_norm": 4.074622631072998,
|
| 1821 |
+
"learning_rate": 6.024841428293787e-07,
|
| 1822 |
+
"loss": 0.1201,
|
| 1823 |
+
"step": 113000
|
| 1824 |
+
},
|
| 1825 |
+
{
|
| 1826 |
+
"epoch": 6.0,
|
| 1827 |
+
"step": 113298,
|
| 1828 |
+
"total_flos": 1.5167000684003328e+18,
|
| 1829 |
+
"train_loss": 0.20791157794172538,
|
| 1830 |
+
"train_runtime": 110416.4573,
|
| 1831 |
+
"train_samples_per_second": 2.052,
|
| 1832 |
+
"train_steps_per_second": 1.026
|
| 1833 |
+
}
|
| 1834 |
+
],
|
| 1835 |
+
"logging_steps": 500,
|
| 1836 |
+
"max_steps": 113298,
|
| 1837 |
+
"num_input_tokens_seen": 0,
|
| 1838 |
+
"num_train_epochs": 6,
|
| 1839 |
+
"save_steps": 500,
|
| 1840 |
+
"stateful_callbacks": {
|
| 1841 |
+
"TrainerControl": {
|
| 1842 |
+
"args": {
|
| 1843 |
+
"should_epoch_stop": false,
|
| 1844 |
+
"should_evaluate": false,
|
| 1845 |
+
"should_log": false,
|
| 1846 |
+
"should_save": true,
|
| 1847 |
+
"should_training_stop": true
|
| 1848 |
+
},
|
| 1849 |
+
"attributes": {}
|
| 1850 |
+
}
|
| 1851 |
+
},
|
| 1852 |
+
"total_flos": 1.5167000684003328e+18,
|
| 1853 |
+
"train_batch_size": 2,
|
| 1854 |
+
"trial_name": null,
|
| 1855 |
+
"trial_params": null
|
| 1856 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:426da0af5a60854e8666dfd7a9f7df351a0428737e05f7c6016ef0ab07256af1
|
| 3 |
+
size 5432
|