Upload 16 files
Browse files- README.md +56 -3
- all_results.json +8 -0
- checkpoint-102000/config.json +55 -0
- checkpoint-102000/model.safetensors +3 -0
- checkpoint-102000/optimizer.pt +3 -0
- checkpoint-102000/preprocessor_config.json +36 -0
- checkpoint-102000/rng_state.pth +3 -0
- checkpoint-102000/scheduler.pt +3 -0
- checkpoint-102000/trainer_state.json +1449 -0
- checkpoint-102000/training_args.bin +3 -0
- config.json +55 -0
- model.safetensors +3 -0
- preprocessor_config.json +36 -0
- train_results.json +8 -0
- trainer_state.json +1458 -0
- training_args.bin +3 -0
README.md
CHANGED
|
@@ -1,3 +1,56 @@
|
|
| 1 |
-
---
|
| 2 |
-
license:
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: google/vit-base-patch16-224-in21k
|
| 4 |
+
tags:
|
| 5 |
+
- image-classification
|
| 6 |
+
- vision
|
| 7 |
+
- generated_from_trainer
|
| 8 |
+
datasets:
|
| 9 |
+
- imagefolder
|
| 10 |
+
model-index:
|
| 11 |
+
- name: network
|
| 12 |
+
results: []
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 16 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 17 |
+
|
| 18 |
+
# network
|
| 19 |
+
|
| 20 |
+
This model is a fine-tuned version of [google/vit-base-patch16-224-in21k](https://huggingface.co/google/vit-base-patch16-224-in21k) on the imagefolder dataset.
|
| 21 |
+
|
| 22 |
+
## Model description
|
| 23 |
+
|
| 24 |
+
More information needed
|
| 25 |
+
|
| 26 |
+
## Intended uses & limitations
|
| 27 |
+
|
| 28 |
+
More information needed
|
| 29 |
+
|
| 30 |
+
## Training and evaluation data
|
| 31 |
+
|
| 32 |
+
More information needed
|
| 33 |
+
|
| 34 |
+
## Training procedure
|
| 35 |
+
|
| 36 |
+
### Training hyperparameters
|
| 37 |
+
|
| 38 |
+
The following hyperparameters were used during training:
|
| 39 |
+
- learning_rate: 5e-05
|
| 40 |
+
- train_batch_size: 8
|
| 41 |
+
- eval_batch_size: 8
|
| 42 |
+
- seed: 42
|
| 43 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 44 |
+
- lr_scheduler_type: linear
|
| 45 |
+
- num_epochs: 3.0
|
| 46 |
+
|
| 47 |
+
### Training results
|
| 48 |
+
|
| 49 |
+
|
| 50 |
+
|
| 51 |
+
### Framework versions
|
| 52 |
+
|
| 53 |
+
- Transformers 4.40.2
|
| 54 |
+
- Pytorch 2.3.0+cu121
|
| 55 |
+
- Datasets 2.19.1
|
| 56 |
+
- Tokenizers 0.19.1
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 6.323969768236646e+19,
|
| 4 |
+
"train_loss": 0.0325347986571929,
|
| 5 |
+
"train_runtime": 10422.0944,
|
| 6 |
+
"train_samples_per_second": 78.295,
|
| 7 |
+
"train_steps_per_second": 9.787
|
| 8 |
+
}
|
checkpoint-102000/config.json
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google/vit-base-patch16-224-in21k",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"ViTForImageClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"encoder_stride": 16,
|
| 8 |
+
"finetuning_task": "image-classification",
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.0,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "bb",
|
| 14 |
+
"1": "bk",
|
| 15 |
+
"10": "wp",
|
| 16 |
+
"11": "wq",
|
| 17 |
+
"12": "wr",
|
| 18 |
+
"2": "bn",
|
| 19 |
+
"3": "bp",
|
| 20 |
+
"4": "bq",
|
| 21 |
+
"5": "br",
|
| 22 |
+
"6": "empty",
|
| 23 |
+
"7": "wb",
|
| 24 |
+
"8": "wk",
|
| 25 |
+
"9": "wn"
|
| 26 |
+
},
|
| 27 |
+
"image_size": 224,
|
| 28 |
+
"initializer_range": 0.02,
|
| 29 |
+
"intermediate_size": 3072,
|
| 30 |
+
"label2id": {
|
| 31 |
+
"bb": "0",
|
| 32 |
+
"bk": "1",
|
| 33 |
+
"bn": "2",
|
| 34 |
+
"bp": "3",
|
| 35 |
+
"bq": "4",
|
| 36 |
+
"br": "5",
|
| 37 |
+
"empty": "6",
|
| 38 |
+
"wb": "7",
|
| 39 |
+
"wk": "8",
|
| 40 |
+
"wn": "9",
|
| 41 |
+
"wp": "10",
|
| 42 |
+
"wq": "11",
|
| 43 |
+
"wr": "12"
|
| 44 |
+
},
|
| 45 |
+
"layer_norm_eps": 1e-12,
|
| 46 |
+
"model_type": "vit",
|
| 47 |
+
"num_attention_heads": 12,
|
| 48 |
+
"num_channels": 3,
|
| 49 |
+
"num_hidden_layers": 12,
|
| 50 |
+
"patch_size": 16,
|
| 51 |
+
"problem_type": "single_label_classification",
|
| 52 |
+
"qkv_bias": true,
|
| 53 |
+
"torch_dtype": "float32",
|
| 54 |
+
"transformers_version": "4.40.2"
|
| 55 |
+
}
|
checkpoint-102000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0fed5d30904017d84562df1d3f165e9382aa1a8c48564aa75d45eed1462f356e
|
| 3 |
+
size 343257812
|
checkpoint-102000/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b694b986d67cad49f8e40b98b5447aee2746498ecd58eadabab7f54e84231027
|
| 3 |
+
size 686636474
|
checkpoint-102000/preprocessor_config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_valid_processor_keys": [
|
| 3 |
+
"images",
|
| 4 |
+
"do_resize",
|
| 5 |
+
"size",
|
| 6 |
+
"resample",
|
| 7 |
+
"do_rescale",
|
| 8 |
+
"rescale_factor",
|
| 9 |
+
"do_normalize",
|
| 10 |
+
"image_mean",
|
| 11 |
+
"image_std",
|
| 12 |
+
"return_tensors",
|
| 13 |
+
"data_format",
|
| 14 |
+
"input_data_format"
|
| 15 |
+
],
|
| 16 |
+
"do_normalize": true,
|
| 17 |
+
"do_rescale": true,
|
| 18 |
+
"do_resize": true,
|
| 19 |
+
"image_mean": [
|
| 20 |
+
0.5,
|
| 21 |
+
0.5,
|
| 22 |
+
0.5
|
| 23 |
+
],
|
| 24 |
+
"image_processor_type": "ViTImageProcessor",
|
| 25 |
+
"image_std": [
|
| 26 |
+
0.5,
|
| 27 |
+
0.5,
|
| 28 |
+
0.5
|
| 29 |
+
],
|
| 30 |
+
"resample": 2,
|
| 31 |
+
"rescale_factor": 0.00392156862745098,
|
| 32 |
+
"size": {
|
| 33 |
+
"height": 224,
|
| 34 |
+
"width": 224
|
| 35 |
+
}
|
| 36 |
+
}
|
checkpoint-102000/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9ee8a41d6942d75a9e923d4b9f35c51b16b26978b7849dbee21a63af2ec0308c
|
| 3 |
+
size 14244
|
checkpoint-102000/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6a15cc9e02c1125fc853e1f62c577b38dbfb76d8a16bcf17f4c7a7f5ce95ee14
|
| 3 |
+
size 1064
|
checkpoint-102000/trainer_state.json
ADDED
|
@@ -0,0 +1,1449 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 102000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.014705882352941176,
|
| 13 |
+
"grad_norm": 0.35731062293052673,
|
| 14 |
+
"learning_rate": 4.975490196078432e-05,
|
| 15 |
+
"loss": 0.5354,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.029411764705882353,
|
| 20 |
+
"grad_norm": 2.1595866680145264,
|
| 21 |
+
"learning_rate": 4.9509803921568634e-05,
|
| 22 |
+
"loss": 0.1627,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.04411764705882353,
|
| 27 |
+
"grad_norm": 0.14825384318828583,
|
| 28 |
+
"learning_rate": 4.9264705882352944e-05,
|
| 29 |
+
"loss": 0.101,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.058823529411764705,
|
| 34 |
+
"grad_norm": 0.07121703773736954,
|
| 35 |
+
"learning_rate": 4.901960784313725e-05,
|
| 36 |
+
"loss": 0.0952,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.07352941176470588,
|
| 41 |
+
"grad_norm": 0.06089532747864723,
|
| 42 |
+
"learning_rate": 4.877450980392157e-05,
|
| 43 |
+
"loss": 0.0881,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.08823529411764706,
|
| 48 |
+
"grad_norm": 0.037034619599580765,
|
| 49 |
+
"learning_rate": 4.8529411764705885e-05,
|
| 50 |
+
"loss": 0.0717,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.10294117647058823,
|
| 55 |
+
"grad_norm": 0.01999847963452339,
|
| 56 |
+
"learning_rate": 4.82843137254902e-05,
|
| 57 |
+
"loss": 0.0848,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.11764705882352941,
|
| 62 |
+
"grad_norm": 0.018650399520993233,
|
| 63 |
+
"learning_rate": 4.803921568627452e-05,
|
| 64 |
+
"loss": 0.068,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.1323529411764706,
|
| 69 |
+
"grad_norm": 0.023023229092359543,
|
| 70 |
+
"learning_rate": 4.7794117647058826e-05,
|
| 71 |
+
"loss": 0.0626,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.14705882352941177,
|
| 76 |
+
"grad_norm": 0.018715515732765198,
|
| 77 |
+
"learning_rate": 4.7549019607843135e-05,
|
| 78 |
+
"loss": 0.0555,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.16176470588235295,
|
| 83 |
+
"grad_norm": 0.11842140555381775,
|
| 84 |
+
"learning_rate": 4.730392156862745e-05,
|
| 85 |
+
"loss": 0.0658,
|
| 86 |
+
"step": 5500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.17647058823529413,
|
| 90 |
+
"grad_norm": 0.04816881939768791,
|
| 91 |
+
"learning_rate": 4.705882352941177e-05,
|
| 92 |
+
"loss": 0.0648,
|
| 93 |
+
"step": 6000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.19117647058823528,
|
| 97 |
+
"grad_norm": 0.025240018963813782,
|
| 98 |
+
"learning_rate": 4.681372549019608e-05,
|
| 99 |
+
"loss": 0.0632,
|
| 100 |
+
"step": 6500
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.20588235294117646,
|
| 104 |
+
"grad_norm": 0.007390766404569149,
|
| 105 |
+
"learning_rate": 4.656862745098039e-05,
|
| 106 |
+
"loss": 0.0705,
|
| 107 |
+
"step": 7000
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.22058823529411764,
|
| 111 |
+
"grad_norm": 0.011664963327348232,
|
| 112 |
+
"learning_rate": 4.632352941176471e-05,
|
| 113 |
+
"loss": 0.0552,
|
| 114 |
+
"step": 7500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.23529411764705882,
|
| 118 |
+
"grad_norm": 39.69175720214844,
|
| 119 |
+
"learning_rate": 4.607843137254902e-05,
|
| 120 |
+
"loss": 0.0686,
|
| 121 |
+
"step": 8000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.25,
|
| 125 |
+
"grad_norm": 0.007366931065917015,
|
| 126 |
+
"learning_rate": 4.5833333333333334e-05,
|
| 127 |
+
"loss": 0.061,
|
| 128 |
+
"step": 8500
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.2647058823529412,
|
| 132 |
+
"grad_norm": 0.003396671498194337,
|
| 133 |
+
"learning_rate": 4.558823529411765e-05,
|
| 134 |
+
"loss": 0.0464,
|
| 135 |
+
"step": 9000
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.27941176470588236,
|
| 139 |
+
"grad_norm": 0.045366521924734116,
|
| 140 |
+
"learning_rate": 4.5343137254901966e-05,
|
| 141 |
+
"loss": 0.053,
|
| 142 |
+
"step": 9500
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.29411764705882354,
|
| 146 |
+
"grad_norm": 1.137495756149292,
|
| 147 |
+
"learning_rate": 4.5098039215686275e-05,
|
| 148 |
+
"loss": 0.0484,
|
| 149 |
+
"step": 10000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.3088235294117647,
|
| 153 |
+
"grad_norm": 0.0062417215667665005,
|
| 154 |
+
"learning_rate": 4.485294117647059e-05,
|
| 155 |
+
"loss": 0.0564,
|
| 156 |
+
"step": 10500
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.3235294117647059,
|
| 160 |
+
"grad_norm": 0.003990447614341974,
|
| 161 |
+
"learning_rate": 4.460784313725491e-05,
|
| 162 |
+
"loss": 0.0364,
|
| 163 |
+
"step": 11000
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.3382352941176471,
|
| 167 |
+
"grad_norm": 0.11569799482822418,
|
| 168 |
+
"learning_rate": 4.4362745098039216e-05,
|
| 169 |
+
"loss": 0.0436,
|
| 170 |
+
"step": 11500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.35294117647058826,
|
| 174 |
+
"grad_norm": 0.01159907691180706,
|
| 175 |
+
"learning_rate": 4.411764705882353e-05,
|
| 176 |
+
"loss": 0.0456,
|
| 177 |
+
"step": 12000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.36764705882352944,
|
| 181 |
+
"grad_norm": 10.926911354064941,
|
| 182 |
+
"learning_rate": 4.387254901960784e-05,
|
| 183 |
+
"loss": 0.0464,
|
| 184 |
+
"step": 12500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.38235294117647056,
|
| 188 |
+
"grad_norm": 0.03444543853402138,
|
| 189 |
+
"learning_rate": 4.362745098039216e-05,
|
| 190 |
+
"loss": 0.0512,
|
| 191 |
+
"step": 13000
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.39705882352941174,
|
| 195 |
+
"grad_norm": 0.002395658055320382,
|
| 196 |
+
"learning_rate": 4.3382352941176474e-05,
|
| 197 |
+
"loss": 0.0586,
|
| 198 |
+
"step": 13500
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.4117647058823529,
|
| 202 |
+
"grad_norm": 0.010224021971225739,
|
| 203 |
+
"learning_rate": 4.313725490196079e-05,
|
| 204 |
+
"loss": 0.0407,
|
| 205 |
+
"step": 14000
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.4264705882352941,
|
| 209 |
+
"grad_norm": 0.010642035864293575,
|
| 210 |
+
"learning_rate": 4.28921568627451e-05,
|
| 211 |
+
"loss": 0.0536,
|
| 212 |
+
"step": 14500
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.4411764705882353,
|
| 216 |
+
"grad_norm": 0.00970557238906622,
|
| 217 |
+
"learning_rate": 4.2647058823529415e-05,
|
| 218 |
+
"loss": 0.0526,
|
| 219 |
+
"step": 15000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.45588235294117646,
|
| 223 |
+
"grad_norm": 0.0030253385193645954,
|
| 224 |
+
"learning_rate": 4.2401960784313724e-05,
|
| 225 |
+
"loss": 0.0432,
|
| 226 |
+
"step": 15500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.47058823529411764,
|
| 230 |
+
"grad_norm": 0.5423064827919006,
|
| 231 |
+
"learning_rate": 4.215686274509804e-05,
|
| 232 |
+
"loss": 0.0381,
|
| 233 |
+
"step": 16000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.4852941176470588,
|
| 237 |
+
"grad_norm": 7.454125881195068,
|
| 238 |
+
"learning_rate": 4.1911764705882356e-05,
|
| 239 |
+
"loss": 0.0397,
|
| 240 |
+
"step": 16500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.5,
|
| 244 |
+
"grad_norm": 3.87744402885437,
|
| 245 |
+
"learning_rate": 4.166666666666667e-05,
|
| 246 |
+
"loss": 0.0564,
|
| 247 |
+
"step": 17000
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.5147058823529411,
|
| 251 |
+
"grad_norm": 33.39067840576172,
|
| 252 |
+
"learning_rate": 4.142156862745099e-05,
|
| 253 |
+
"loss": 0.0455,
|
| 254 |
+
"step": 17500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.5294117647058824,
|
| 258 |
+
"grad_norm": 0.008246080949902534,
|
| 259 |
+
"learning_rate": 4.11764705882353e-05,
|
| 260 |
+
"loss": 0.0515,
|
| 261 |
+
"step": 18000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.5441176470588235,
|
| 265 |
+
"grad_norm": 0.018734918907284737,
|
| 266 |
+
"learning_rate": 4.0931372549019607e-05,
|
| 267 |
+
"loss": 0.0453,
|
| 268 |
+
"step": 18500
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.5588235294117647,
|
| 272 |
+
"grad_norm": 0.007373027969151735,
|
| 273 |
+
"learning_rate": 4.068627450980392e-05,
|
| 274 |
+
"loss": 0.0485,
|
| 275 |
+
"step": 19000
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.5735294117647058,
|
| 279 |
+
"grad_norm": 0.0023267469368875027,
|
| 280 |
+
"learning_rate": 4.044117647058824e-05,
|
| 281 |
+
"loss": 0.0372,
|
| 282 |
+
"step": 19500
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.5882352941176471,
|
| 286 |
+
"grad_norm": 0.1311252862215042,
|
| 287 |
+
"learning_rate": 4.0196078431372555e-05,
|
| 288 |
+
"loss": 0.0442,
|
| 289 |
+
"step": 20000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.6029411764705882,
|
| 293 |
+
"grad_norm": 0.003710985416546464,
|
| 294 |
+
"learning_rate": 3.9950980392156864e-05,
|
| 295 |
+
"loss": 0.0585,
|
| 296 |
+
"step": 20500
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.6176470588235294,
|
| 300 |
+
"grad_norm": 0.004861747846007347,
|
| 301 |
+
"learning_rate": 3.970588235294117e-05,
|
| 302 |
+
"loss": 0.0374,
|
| 303 |
+
"step": 21000
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.6323529411764706,
|
| 307 |
+
"grad_norm": 0.004351571202278137,
|
| 308 |
+
"learning_rate": 3.946078431372549e-05,
|
| 309 |
+
"loss": 0.0265,
|
| 310 |
+
"step": 21500
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.6470588235294118,
|
| 314 |
+
"grad_norm": 0.008851751685142517,
|
| 315 |
+
"learning_rate": 3.9215686274509805e-05,
|
| 316 |
+
"loss": 0.0314,
|
| 317 |
+
"step": 22000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.6617647058823529,
|
| 321 |
+
"grad_norm": 0.0046307104639709,
|
| 322 |
+
"learning_rate": 3.897058823529412e-05,
|
| 323 |
+
"loss": 0.0445,
|
| 324 |
+
"step": 22500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.6764705882352942,
|
| 328 |
+
"grad_norm": 0.002735880669206381,
|
| 329 |
+
"learning_rate": 3.872549019607844e-05,
|
| 330 |
+
"loss": 0.0433,
|
| 331 |
+
"step": 23000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.6911764705882353,
|
| 335 |
+
"grad_norm": 0.0218490082770586,
|
| 336 |
+
"learning_rate": 3.8480392156862746e-05,
|
| 337 |
+
"loss": 0.0444,
|
| 338 |
+
"step": 23500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.7058823529411765,
|
| 342 |
+
"grad_norm": 2.371448278427124,
|
| 343 |
+
"learning_rate": 3.8235294117647055e-05,
|
| 344 |
+
"loss": 0.033,
|
| 345 |
+
"step": 24000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.7205882352941176,
|
| 349 |
+
"grad_norm": 0.005015780217945576,
|
| 350 |
+
"learning_rate": 3.799019607843137e-05,
|
| 351 |
+
"loss": 0.0422,
|
| 352 |
+
"step": 24500
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.7352941176470589,
|
| 356 |
+
"grad_norm": 1.3102850914001465,
|
| 357 |
+
"learning_rate": 3.774509803921569e-05,
|
| 358 |
+
"loss": 0.035,
|
| 359 |
+
"step": 25000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.75,
|
| 363 |
+
"grad_norm": 0.004522955510765314,
|
| 364 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 365 |
+
"loss": 0.0303,
|
| 366 |
+
"step": 25500
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.7647058823529411,
|
| 370 |
+
"grad_norm": 0.006077844649553299,
|
| 371 |
+
"learning_rate": 3.725490196078432e-05,
|
| 372 |
+
"loss": 0.0503,
|
| 373 |
+
"step": 26000
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.7794117647058824,
|
| 377 |
+
"grad_norm": 0.0026446939446032047,
|
| 378 |
+
"learning_rate": 3.700980392156863e-05,
|
| 379 |
+
"loss": 0.0264,
|
| 380 |
+
"step": 26500
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.7941176470588235,
|
| 384 |
+
"grad_norm": 0.004485867917537689,
|
| 385 |
+
"learning_rate": 3.6764705882352945e-05,
|
| 386 |
+
"loss": 0.0388,
|
| 387 |
+
"step": 27000
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.8088235294117647,
|
| 391 |
+
"grad_norm": 0.0073866695165634155,
|
| 392 |
+
"learning_rate": 3.6519607843137254e-05,
|
| 393 |
+
"loss": 0.0283,
|
| 394 |
+
"step": 27500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.8235294117647058,
|
| 398 |
+
"grad_norm": 0.012984287925064564,
|
| 399 |
+
"learning_rate": 3.627450980392157e-05,
|
| 400 |
+
"loss": 0.0393,
|
| 401 |
+
"step": 28000
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.8382352941176471,
|
| 405 |
+
"grad_norm": 0.01751883700489998,
|
| 406 |
+
"learning_rate": 3.6029411764705886e-05,
|
| 407 |
+
"loss": 0.0388,
|
| 408 |
+
"step": 28500
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.8529411764705882,
|
| 412 |
+
"grad_norm": 0.0035843336954712868,
|
| 413 |
+
"learning_rate": 3.5784313725490195e-05,
|
| 414 |
+
"loss": 0.0384,
|
| 415 |
+
"step": 29000
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.8676470588235294,
|
| 419 |
+
"grad_norm": 0.0645672082901001,
|
| 420 |
+
"learning_rate": 3.553921568627451e-05,
|
| 421 |
+
"loss": 0.0378,
|
| 422 |
+
"step": 29500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.8823529411764706,
|
| 426 |
+
"grad_norm": 0.008566264994442463,
|
| 427 |
+
"learning_rate": 3.529411764705883e-05,
|
| 428 |
+
"loss": 0.0315,
|
| 429 |
+
"step": 30000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.8970588235294118,
|
| 433 |
+
"grad_norm": 0.010571740567684174,
|
| 434 |
+
"learning_rate": 3.5049019607843136e-05,
|
| 435 |
+
"loss": 0.0324,
|
| 436 |
+
"step": 30500
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.9117647058823529,
|
| 440 |
+
"grad_norm": 0.0022533361334353685,
|
| 441 |
+
"learning_rate": 3.480392156862745e-05,
|
| 442 |
+
"loss": 0.0292,
|
| 443 |
+
"step": 31000
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.9264705882352942,
|
| 447 |
+
"grad_norm": 0.006164130289107561,
|
| 448 |
+
"learning_rate": 3.455882352941177e-05,
|
| 449 |
+
"loss": 0.0407,
|
| 450 |
+
"step": 31500
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.9411764705882353,
|
| 454 |
+
"grad_norm": 0.007435985840857029,
|
| 455 |
+
"learning_rate": 3.431372549019608e-05,
|
| 456 |
+
"loss": 0.0449,
|
| 457 |
+
"step": 32000
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.9558823529411765,
|
| 461 |
+
"grad_norm": 0.003777585458010435,
|
| 462 |
+
"learning_rate": 3.4068627450980394e-05,
|
| 463 |
+
"loss": 0.0247,
|
| 464 |
+
"step": 32500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.9705882352941176,
|
| 468 |
+
"grad_norm": 0.005975374951958656,
|
| 469 |
+
"learning_rate": 3.382352941176471e-05,
|
| 470 |
+
"loss": 0.0309,
|
| 471 |
+
"step": 33000
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.9852941176470589,
|
| 475 |
+
"grad_norm": 0.0016012012492865324,
|
| 476 |
+
"learning_rate": 3.357843137254902e-05,
|
| 477 |
+
"loss": 0.0294,
|
| 478 |
+
"step": 33500
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.0,
|
| 482 |
+
"grad_norm": 0.020027656108140945,
|
| 483 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 484 |
+
"loss": 0.022,
|
| 485 |
+
"step": 34000
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.0147058823529411,
|
| 489 |
+
"grad_norm": 10.916418075561523,
|
| 490 |
+
"learning_rate": 3.308823529411765e-05,
|
| 491 |
+
"loss": 0.0264,
|
| 492 |
+
"step": 34500
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.0294117647058822,
|
| 496 |
+
"grad_norm": 0.013728056102991104,
|
| 497 |
+
"learning_rate": 3.284313725490196e-05,
|
| 498 |
+
"loss": 0.032,
|
| 499 |
+
"step": 35000
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.0441176470588236,
|
| 503 |
+
"grad_norm": 0.02546643279492855,
|
| 504 |
+
"learning_rate": 3.2598039215686276e-05,
|
| 505 |
+
"loss": 0.0404,
|
| 506 |
+
"step": 35500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.0588235294117647,
|
| 510 |
+
"grad_norm": 0.004185052588582039,
|
| 511 |
+
"learning_rate": 3.235294117647059e-05,
|
| 512 |
+
"loss": 0.0306,
|
| 513 |
+
"step": 36000
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.0735294117647058,
|
| 517 |
+
"grad_norm": 0.016653403639793396,
|
| 518 |
+
"learning_rate": 3.210784313725491e-05,
|
| 519 |
+
"loss": 0.0217,
|
| 520 |
+
"step": 36500
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.088235294117647,
|
| 524 |
+
"grad_norm": 0.0028331661596894264,
|
| 525 |
+
"learning_rate": 3.186274509803922e-05,
|
| 526 |
+
"loss": 0.0265,
|
| 527 |
+
"step": 37000
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.1029411764705883,
|
| 531 |
+
"grad_norm": 0.013931985944509506,
|
| 532 |
+
"learning_rate": 3.161764705882353e-05,
|
| 533 |
+
"loss": 0.0399,
|
| 534 |
+
"step": 37500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.1176470588235294,
|
| 538 |
+
"grad_norm": 0.001668413169682026,
|
| 539 |
+
"learning_rate": 3.137254901960784e-05,
|
| 540 |
+
"loss": 0.0202,
|
| 541 |
+
"step": 38000
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.1323529411764706,
|
| 545 |
+
"grad_norm": 0.00567347789183259,
|
| 546 |
+
"learning_rate": 3.112745098039216e-05,
|
| 547 |
+
"loss": 0.0336,
|
| 548 |
+
"step": 38500
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.1470588235294117,
|
| 552 |
+
"grad_norm": 0.003477458842098713,
|
| 553 |
+
"learning_rate": 3.0882352941176475e-05,
|
| 554 |
+
"loss": 0.0253,
|
| 555 |
+
"step": 39000
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.161764705882353,
|
| 559 |
+
"grad_norm": 0.21819466352462769,
|
| 560 |
+
"learning_rate": 3.063725490196079e-05,
|
| 561 |
+
"loss": 0.0293,
|
| 562 |
+
"step": 39500
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.1764705882352942,
|
| 566 |
+
"grad_norm": 0.004708552733063698,
|
| 567 |
+
"learning_rate": 3.0392156862745097e-05,
|
| 568 |
+
"loss": 0.0268,
|
| 569 |
+
"step": 40000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.1911764705882353,
|
| 573 |
+
"grad_norm": 0.0018363581039011478,
|
| 574 |
+
"learning_rate": 3.0147058823529413e-05,
|
| 575 |
+
"loss": 0.0266,
|
| 576 |
+
"step": 40500
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.2058823529411764,
|
| 580 |
+
"grad_norm": 0.005581580102443695,
|
| 581 |
+
"learning_rate": 2.9901960784313725e-05,
|
| 582 |
+
"loss": 0.0361,
|
| 583 |
+
"step": 41000
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.2205882352941178,
|
| 587 |
+
"grad_norm": 0.0035322746261954308,
|
| 588 |
+
"learning_rate": 2.965686274509804e-05,
|
| 589 |
+
"loss": 0.0274,
|
| 590 |
+
"step": 41500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.2352941176470589,
|
| 594 |
+
"grad_norm": 0.0038708180654793978,
|
| 595 |
+
"learning_rate": 2.9411764705882354e-05,
|
| 596 |
+
"loss": 0.0241,
|
| 597 |
+
"step": 42000
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.25,
|
| 601 |
+
"grad_norm": 0.0019606975838541985,
|
| 602 |
+
"learning_rate": 2.916666666666667e-05,
|
| 603 |
+
"loss": 0.0219,
|
| 604 |
+
"step": 42500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.2647058823529411,
|
| 608 |
+
"grad_norm": 0.00241417670622468,
|
| 609 |
+
"learning_rate": 2.8921568627450986e-05,
|
| 610 |
+
"loss": 0.0244,
|
| 611 |
+
"step": 43000
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.2794117647058822,
|
| 615 |
+
"grad_norm": 0.003593308152630925,
|
| 616 |
+
"learning_rate": 2.8676470588235295e-05,
|
| 617 |
+
"loss": 0.027,
|
| 618 |
+
"step": 43500
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.2941176470588236,
|
| 622 |
+
"grad_norm": 1.4003372192382812,
|
| 623 |
+
"learning_rate": 2.8431372549019608e-05,
|
| 624 |
+
"loss": 0.0392,
|
| 625 |
+
"step": 44000
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.3088235294117647,
|
| 629 |
+
"grad_norm": 0.41921645402908325,
|
| 630 |
+
"learning_rate": 2.8186274509803924e-05,
|
| 631 |
+
"loss": 0.0299,
|
| 632 |
+
"step": 44500
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.3235294117647058,
|
| 636 |
+
"grad_norm": 1.567896842956543,
|
| 637 |
+
"learning_rate": 2.7941176470588236e-05,
|
| 638 |
+
"loss": 0.0282,
|
| 639 |
+
"step": 45000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.3382352941176472,
|
| 643 |
+
"grad_norm": 0.003143745008856058,
|
| 644 |
+
"learning_rate": 2.7696078431372552e-05,
|
| 645 |
+
"loss": 0.0197,
|
| 646 |
+
"step": 45500
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.3529411764705883,
|
| 650 |
+
"grad_norm": 0.0016428233357146382,
|
| 651 |
+
"learning_rate": 2.7450980392156865e-05,
|
| 652 |
+
"loss": 0.0344,
|
| 653 |
+
"step": 46000
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.3676470588235294,
|
| 657 |
+
"grad_norm": 0.005276167765259743,
|
| 658 |
+
"learning_rate": 2.7205882352941174e-05,
|
| 659 |
+
"loss": 0.0312,
|
| 660 |
+
"step": 46500
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.3823529411764706,
|
| 664 |
+
"grad_norm": 0.004461635369807482,
|
| 665 |
+
"learning_rate": 2.696078431372549e-05,
|
| 666 |
+
"loss": 0.025,
|
| 667 |
+
"step": 47000
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.3970588235294117,
|
| 671 |
+
"grad_norm": 0.003238542238250375,
|
| 672 |
+
"learning_rate": 2.6715686274509806e-05,
|
| 673 |
+
"loss": 0.0236,
|
| 674 |
+
"step": 47500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.4117647058823528,
|
| 678 |
+
"grad_norm": 0.0017602238804101944,
|
| 679 |
+
"learning_rate": 2.647058823529412e-05,
|
| 680 |
+
"loss": 0.0234,
|
| 681 |
+
"step": 48000
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.4264705882352942,
|
| 685 |
+
"grad_norm": 0.002768098609521985,
|
| 686 |
+
"learning_rate": 2.6225490196078435e-05,
|
| 687 |
+
"loss": 0.026,
|
| 688 |
+
"step": 48500
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.4411764705882353,
|
| 692 |
+
"grad_norm": 0.002579926745966077,
|
| 693 |
+
"learning_rate": 2.5980392156862747e-05,
|
| 694 |
+
"loss": 0.0167,
|
| 695 |
+
"step": 49000
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.4558823529411764,
|
| 699 |
+
"grad_norm": 0.008888750337064266,
|
| 700 |
+
"learning_rate": 2.5735294117647057e-05,
|
| 701 |
+
"loss": 0.0288,
|
| 702 |
+
"step": 49500
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.4705882352941178,
|
| 706 |
+
"grad_norm": 0.0038515792693942785,
|
| 707 |
+
"learning_rate": 2.5490196078431373e-05,
|
| 708 |
+
"loss": 0.0252,
|
| 709 |
+
"step": 50000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.4852941176470589,
|
| 713 |
+
"grad_norm": 4.428643226623535,
|
| 714 |
+
"learning_rate": 2.5245098039215685e-05,
|
| 715 |
+
"loss": 0.016,
|
| 716 |
+
"step": 50500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.5,
|
| 720 |
+
"grad_norm": 0.002629584399983287,
|
| 721 |
+
"learning_rate": 2.5e-05,
|
| 722 |
+
"loss": 0.0214,
|
| 723 |
+
"step": 51000
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 1.5147058823529411,
|
| 727 |
+
"grad_norm": 0.004097859375178814,
|
| 728 |
+
"learning_rate": 2.4754901960784317e-05,
|
| 729 |
+
"loss": 0.0183,
|
| 730 |
+
"step": 51500
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.5294117647058822,
|
| 734 |
+
"grad_norm": 0.00568019924685359,
|
| 735 |
+
"learning_rate": 2.4509803921568626e-05,
|
| 736 |
+
"loss": 0.0271,
|
| 737 |
+
"step": 52000
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.5441176470588234,
|
| 741 |
+
"grad_norm": 0.0020534582436084747,
|
| 742 |
+
"learning_rate": 2.4264705882352942e-05,
|
| 743 |
+
"loss": 0.0281,
|
| 744 |
+
"step": 52500
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.5588235294117647,
|
| 748 |
+
"grad_norm": 0.003300599753856659,
|
| 749 |
+
"learning_rate": 2.401960784313726e-05,
|
| 750 |
+
"loss": 0.0284,
|
| 751 |
+
"step": 53000
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.5735294117647058,
|
| 755 |
+
"grad_norm": 0.001472037984058261,
|
| 756 |
+
"learning_rate": 2.3774509803921568e-05,
|
| 757 |
+
"loss": 0.0205,
|
| 758 |
+
"step": 53500
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.5882352941176472,
|
| 762 |
+
"grad_norm": 0.002835978288203478,
|
| 763 |
+
"learning_rate": 2.3529411764705884e-05,
|
| 764 |
+
"loss": 0.0234,
|
| 765 |
+
"step": 54000
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 1.6029411764705883,
|
| 769 |
+
"grad_norm": 0.003979724366217852,
|
| 770 |
+
"learning_rate": 2.3284313725490196e-05,
|
| 771 |
+
"loss": 0.0233,
|
| 772 |
+
"step": 54500
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 1.6176470588235294,
|
| 776 |
+
"grad_norm": 0.0036057273391634226,
|
| 777 |
+
"learning_rate": 2.303921568627451e-05,
|
| 778 |
+
"loss": 0.0319,
|
| 779 |
+
"step": 55000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.6323529411764706,
|
| 783 |
+
"grad_norm": 0.0024822901468724012,
|
| 784 |
+
"learning_rate": 2.2794117647058825e-05,
|
| 785 |
+
"loss": 0.0211,
|
| 786 |
+
"step": 55500
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 1.6470588235294117,
|
| 790 |
+
"grad_norm": 0.008930359967052937,
|
| 791 |
+
"learning_rate": 2.2549019607843138e-05,
|
| 792 |
+
"loss": 0.0309,
|
| 793 |
+
"step": 56000
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.6617647058823528,
|
| 797 |
+
"grad_norm": 1.3381928205490112,
|
| 798 |
+
"learning_rate": 2.2303921568627454e-05,
|
| 799 |
+
"loss": 0.019,
|
| 800 |
+
"step": 56500
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.6764705882352942,
|
| 804 |
+
"grad_norm": 0.004481327719986439,
|
| 805 |
+
"learning_rate": 2.2058823529411766e-05,
|
| 806 |
+
"loss": 0.0228,
|
| 807 |
+
"step": 57000
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 1.6911764705882353,
|
| 811 |
+
"grad_norm": 0.012597435154020786,
|
| 812 |
+
"learning_rate": 2.181372549019608e-05,
|
| 813 |
+
"loss": 0.0285,
|
| 814 |
+
"step": 57500
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.7058823529411766,
|
| 818 |
+
"grad_norm": 0.0034783107694238424,
|
| 819 |
+
"learning_rate": 2.1568627450980395e-05,
|
| 820 |
+
"loss": 0.0165,
|
| 821 |
+
"step": 58000
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 1.7205882352941178,
|
| 825 |
+
"grad_norm": 0.0032083301339298487,
|
| 826 |
+
"learning_rate": 2.1323529411764707e-05,
|
| 827 |
+
"loss": 0.0306,
|
| 828 |
+
"step": 58500
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 1.7352941176470589,
|
| 832 |
+
"grad_norm": 0.0039000194519758224,
|
| 833 |
+
"learning_rate": 2.107843137254902e-05,
|
| 834 |
+
"loss": 0.0199,
|
| 835 |
+
"step": 59000
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 1.75,
|
| 839 |
+
"grad_norm": 0.0034919639583677053,
|
| 840 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 841 |
+
"loss": 0.034,
|
| 842 |
+
"step": 59500
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.7647058823529411,
|
| 846 |
+
"grad_norm": 0.004928025882691145,
|
| 847 |
+
"learning_rate": 2.058823529411765e-05,
|
| 848 |
+
"loss": 0.0213,
|
| 849 |
+
"step": 60000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.7794117647058822,
|
| 853 |
+
"grad_norm": 0.003211489412933588,
|
| 854 |
+
"learning_rate": 2.034313725490196e-05,
|
| 855 |
+
"loss": 0.0206,
|
| 856 |
+
"step": 60500
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.7941176470588234,
|
| 860 |
+
"grad_norm": 0.011239697225391865,
|
| 861 |
+
"learning_rate": 2.0098039215686277e-05,
|
| 862 |
+
"loss": 0.0201,
|
| 863 |
+
"step": 61000
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 1.8088235294117647,
|
| 867 |
+
"grad_norm": 0.0024609589017927647,
|
| 868 |
+
"learning_rate": 1.9852941176470586e-05,
|
| 869 |
+
"loss": 0.0286,
|
| 870 |
+
"step": 61500
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 1.8235294117647058,
|
| 874 |
+
"grad_norm": 0.0046806493774056435,
|
| 875 |
+
"learning_rate": 1.9607843137254903e-05,
|
| 876 |
+
"loss": 0.0216,
|
| 877 |
+
"step": 62000
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 1.8382352941176472,
|
| 881 |
+
"grad_norm": 2.30717396736145,
|
| 882 |
+
"learning_rate": 1.936274509803922e-05,
|
| 883 |
+
"loss": 0.0246,
|
| 884 |
+
"step": 62500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 1.8529411764705883,
|
| 888 |
+
"grad_norm": 0.00428669573739171,
|
| 889 |
+
"learning_rate": 1.9117647058823528e-05,
|
| 890 |
+
"loss": 0.0166,
|
| 891 |
+
"step": 63000
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 1.8676470588235294,
|
| 895 |
+
"grad_norm": 0.011403021402657032,
|
| 896 |
+
"learning_rate": 1.8872549019607844e-05,
|
| 897 |
+
"loss": 0.0272,
|
| 898 |
+
"step": 63500
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 1.8823529411764706,
|
| 902 |
+
"grad_norm": 0.0065813250839710236,
|
| 903 |
+
"learning_rate": 1.862745098039216e-05,
|
| 904 |
+
"loss": 0.0217,
|
| 905 |
+
"step": 64000
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 1.8970588235294117,
|
| 909 |
+
"grad_norm": 0.0024323465768247843,
|
| 910 |
+
"learning_rate": 1.8382352941176472e-05,
|
| 911 |
+
"loss": 0.0176,
|
| 912 |
+
"step": 64500
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.9117647058823528,
|
| 916 |
+
"grad_norm": 0.001616469700820744,
|
| 917 |
+
"learning_rate": 1.8137254901960785e-05,
|
| 918 |
+
"loss": 0.0169,
|
| 919 |
+
"step": 65000
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 1.9264705882352942,
|
| 923 |
+
"grad_norm": 0.004322522785514593,
|
| 924 |
+
"learning_rate": 1.7892156862745098e-05,
|
| 925 |
+
"loss": 0.0216,
|
| 926 |
+
"step": 65500
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 1.9411764705882353,
|
| 930 |
+
"grad_norm": 0.0024695000611245632,
|
| 931 |
+
"learning_rate": 1.7647058823529414e-05,
|
| 932 |
+
"loss": 0.032,
|
| 933 |
+
"step": 66000
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.9558823529411766,
|
| 937 |
+
"grad_norm": 0.010675052180886269,
|
| 938 |
+
"learning_rate": 1.7401960784313726e-05,
|
| 939 |
+
"loss": 0.023,
|
| 940 |
+
"step": 66500
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 1.9705882352941178,
|
| 944 |
+
"grad_norm": 0.49467232823371887,
|
| 945 |
+
"learning_rate": 1.715686274509804e-05,
|
| 946 |
+
"loss": 0.0169,
|
| 947 |
+
"step": 67000
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.9852941176470589,
|
| 951 |
+
"grad_norm": 0.01075649168342352,
|
| 952 |
+
"learning_rate": 1.6911764705882355e-05,
|
| 953 |
+
"loss": 0.0177,
|
| 954 |
+
"step": 67500
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 2.0,
|
| 958 |
+
"grad_norm": 0.026703685522079468,
|
| 959 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 960 |
+
"loss": 0.0259,
|
| 961 |
+
"step": 68000
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 2.014705882352941,
|
| 965 |
+
"grad_norm": 0.0023890878073871136,
|
| 966 |
+
"learning_rate": 1.642156862745098e-05,
|
| 967 |
+
"loss": 0.0128,
|
| 968 |
+
"step": 68500
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 2.0294117647058822,
|
| 972 |
+
"grad_norm": 0.0015898487763479352,
|
| 973 |
+
"learning_rate": 1.6176470588235296e-05,
|
| 974 |
+
"loss": 0.0112,
|
| 975 |
+
"step": 69000
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 2.0441176470588234,
|
| 979 |
+
"grad_norm": 0.0041648312471807,
|
| 980 |
+
"learning_rate": 1.593137254901961e-05,
|
| 981 |
+
"loss": 0.0155,
|
| 982 |
+
"step": 69500
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 2.0588235294117645,
|
| 986 |
+
"grad_norm": 0.10951696336269379,
|
| 987 |
+
"learning_rate": 1.568627450980392e-05,
|
| 988 |
+
"loss": 0.0225,
|
| 989 |
+
"step": 70000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 2.073529411764706,
|
| 993 |
+
"grad_norm": 0.0021414640359580517,
|
| 994 |
+
"learning_rate": 1.5441176470588237e-05,
|
| 995 |
+
"loss": 0.0174,
|
| 996 |
+
"step": 70500
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 2.088235294117647,
|
| 1000 |
+
"grad_norm": 0.001627126126550138,
|
| 1001 |
+
"learning_rate": 1.5196078431372548e-05,
|
| 1002 |
+
"loss": 0.0154,
|
| 1003 |
+
"step": 71000
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 2.1029411764705883,
|
| 1007 |
+
"grad_norm": 0.005821748171001673,
|
| 1008 |
+
"learning_rate": 1.4950980392156863e-05,
|
| 1009 |
+
"loss": 0.0255,
|
| 1010 |
+
"step": 71500
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 2.1176470588235294,
|
| 1014 |
+
"grad_norm": 0.0071876379661262035,
|
| 1015 |
+
"learning_rate": 1.4705882352941177e-05,
|
| 1016 |
+
"loss": 0.024,
|
| 1017 |
+
"step": 72000
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 2.1323529411764706,
|
| 1021 |
+
"grad_norm": 0.00294076488353312,
|
| 1022 |
+
"learning_rate": 1.4460784313725493e-05,
|
| 1023 |
+
"loss": 0.012,
|
| 1024 |
+
"step": 72500
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 2.1470588235294117,
|
| 1028 |
+
"grad_norm": 0.004260234069079161,
|
| 1029 |
+
"learning_rate": 1.4215686274509804e-05,
|
| 1030 |
+
"loss": 0.0202,
|
| 1031 |
+
"step": 73000
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 2.161764705882353,
|
| 1035 |
+
"grad_norm": 0.005722737871110439,
|
| 1036 |
+
"learning_rate": 1.3970588235294118e-05,
|
| 1037 |
+
"loss": 0.0283,
|
| 1038 |
+
"step": 73500
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 2.176470588235294,
|
| 1042 |
+
"grad_norm": 0.016162721440196037,
|
| 1043 |
+
"learning_rate": 1.3725490196078432e-05,
|
| 1044 |
+
"loss": 0.0115,
|
| 1045 |
+
"step": 74000
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 2.1911764705882355,
|
| 1049 |
+
"grad_norm": 0.005048350431025028,
|
| 1050 |
+
"learning_rate": 1.3480392156862745e-05,
|
| 1051 |
+
"loss": 0.023,
|
| 1052 |
+
"step": 74500
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 2.2058823529411766,
|
| 1056 |
+
"grad_norm": 0.004102902952581644,
|
| 1057 |
+
"learning_rate": 1.323529411764706e-05,
|
| 1058 |
+
"loss": 0.018,
|
| 1059 |
+
"step": 75000
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 2.2205882352941178,
|
| 1063 |
+
"grad_norm": 0.0021536860149353743,
|
| 1064 |
+
"learning_rate": 1.2990196078431374e-05,
|
| 1065 |
+
"loss": 0.0169,
|
| 1066 |
+
"step": 75500
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 2.235294117647059,
|
| 1070 |
+
"grad_norm": 0.001363063813187182,
|
| 1071 |
+
"learning_rate": 1.2745098039215686e-05,
|
| 1072 |
+
"loss": 0.0129,
|
| 1073 |
+
"step": 76000
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 2.25,
|
| 1077 |
+
"grad_norm": 0.0025944672524929047,
|
| 1078 |
+
"learning_rate": 1.25e-05,
|
| 1079 |
+
"loss": 0.0095,
|
| 1080 |
+
"step": 76500
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 2.264705882352941,
|
| 1084 |
+
"grad_norm": 0.001098868204280734,
|
| 1085 |
+
"learning_rate": 1.2254901960784313e-05,
|
| 1086 |
+
"loss": 0.0183,
|
| 1087 |
+
"step": 77000
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 2.2794117647058822,
|
| 1091 |
+
"grad_norm": 0.00421817135065794,
|
| 1092 |
+
"learning_rate": 1.200980392156863e-05,
|
| 1093 |
+
"loss": 0.0214,
|
| 1094 |
+
"step": 77500
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 2.2941176470588234,
|
| 1098 |
+
"grad_norm": 0.0016445108922198415,
|
| 1099 |
+
"learning_rate": 1.1764705882352942e-05,
|
| 1100 |
+
"loss": 0.009,
|
| 1101 |
+
"step": 78000
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 2.3088235294117645,
|
| 1105 |
+
"grad_norm": 0.002265740418806672,
|
| 1106 |
+
"learning_rate": 1.1519607843137254e-05,
|
| 1107 |
+
"loss": 0.0161,
|
| 1108 |
+
"step": 78500
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 2.323529411764706,
|
| 1112 |
+
"grad_norm": 0.007099386304616928,
|
| 1113 |
+
"learning_rate": 1.1274509803921569e-05,
|
| 1114 |
+
"loss": 0.0143,
|
| 1115 |
+
"step": 79000
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 2.338235294117647,
|
| 1119 |
+
"grad_norm": 0.003914414439350367,
|
| 1120 |
+
"learning_rate": 1.1029411764705883e-05,
|
| 1121 |
+
"loss": 0.0115,
|
| 1122 |
+
"step": 79500
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 2.3529411764705883,
|
| 1126 |
+
"grad_norm": 0.002513893647119403,
|
| 1127 |
+
"learning_rate": 1.0784313725490197e-05,
|
| 1128 |
+
"loss": 0.0136,
|
| 1129 |
+
"step": 80000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 2.3676470588235294,
|
| 1133 |
+
"grad_norm": 0.0027356306090950966,
|
| 1134 |
+
"learning_rate": 1.053921568627451e-05,
|
| 1135 |
+
"loss": 0.0247,
|
| 1136 |
+
"step": 80500
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 2.3823529411764706,
|
| 1140 |
+
"grad_norm": 0.0817839726805687,
|
| 1141 |
+
"learning_rate": 1.0294117647058824e-05,
|
| 1142 |
+
"loss": 0.0169,
|
| 1143 |
+
"step": 81000
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 2.3970588235294117,
|
| 1147 |
+
"grad_norm": 0.08163878321647644,
|
| 1148 |
+
"learning_rate": 1.0049019607843139e-05,
|
| 1149 |
+
"loss": 0.0114,
|
| 1150 |
+
"step": 81500
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 2.411764705882353,
|
| 1154 |
+
"grad_norm": 0.002625884721055627,
|
| 1155 |
+
"learning_rate": 9.803921568627451e-06,
|
| 1156 |
+
"loss": 0.0147,
|
| 1157 |
+
"step": 82000
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 2.426470588235294,
|
| 1161 |
+
"grad_norm": 0.004531237296760082,
|
| 1162 |
+
"learning_rate": 9.558823529411764e-06,
|
| 1163 |
+
"loss": 0.0106,
|
| 1164 |
+
"step": 82500
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 2.4411764705882355,
|
| 1168 |
+
"grad_norm": 0.0032483581453561783,
|
| 1169 |
+
"learning_rate": 9.31372549019608e-06,
|
| 1170 |
+
"loss": 0.0183,
|
| 1171 |
+
"step": 83000
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 2.4558823529411766,
|
| 1175 |
+
"grad_norm": 0.0050782193429768085,
|
| 1176 |
+
"learning_rate": 9.068627450980392e-06,
|
| 1177 |
+
"loss": 0.0168,
|
| 1178 |
+
"step": 83500
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 2.4705882352941178,
|
| 1182 |
+
"grad_norm": 0.003839900717139244,
|
| 1183 |
+
"learning_rate": 8.823529411764707e-06,
|
| 1184 |
+
"loss": 0.0209,
|
| 1185 |
+
"step": 84000
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 2.485294117647059,
|
| 1189 |
+
"grad_norm": 0.008910595439374447,
|
| 1190 |
+
"learning_rate": 8.57843137254902e-06,
|
| 1191 |
+
"loss": 0.0271,
|
| 1192 |
+
"step": 84500
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 2.5,
|
| 1196 |
+
"grad_norm": 0.008313077501952648,
|
| 1197 |
+
"learning_rate": 8.333333333333334e-06,
|
| 1198 |
+
"loss": 0.0175,
|
| 1199 |
+
"step": 85000
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 2.514705882352941,
|
| 1203 |
+
"grad_norm": 0.0029984668362885714,
|
| 1204 |
+
"learning_rate": 8.088235294117648e-06,
|
| 1205 |
+
"loss": 0.0084,
|
| 1206 |
+
"step": 85500
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 2.5294117647058822,
|
| 1210 |
+
"grad_norm": 0.004524989053606987,
|
| 1211 |
+
"learning_rate": 7.84313725490196e-06,
|
| 1212 |
+
"loss": 0.0156,
|
| 1213 |
+
"step": 86000
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 2.5441176470588234,
|
| 1217 |
+
"grad_norm": 0.0023315059952437878,
|
| 1218 |
+
"learning_rate": 7.598039215686274e-06,
|
| 1219 |
+
"loss": 0.0121,
|
| 1220 |
+
"step": 86500
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 2.5588235294117645,
|
| 1224 |
+
"grad_norm": 0.0010077670449391007,
|
| 1225 |
+
"learning_rate": 7.3529411764705884e-06,
|
| 1226 |
+
"loss": 0.0113,
|
| 1227 |
+
"step": 87000
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 2.5735294117647056,
|
| 1231 |
+
"grad_norm": 0.0012684785760939121,
|
| 1232 |
+
"learning_rate": 7.107843137254902e-06,
|
| 1233 |
+
"loss": 0.0204,
|
| 1234 |
+
"step": 87500
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 2.588235294117647,
|
| 1238 |
+
"grad_norm": 0.004918810911476612,
|
| 1239 |
+
"learning_rate": 6.862745098039216e-06,
|
| 1240 |
+
"loss": 0.0059,
|
| 1241 |
+
"step": 88000
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 2.6029411764705883,
|
| 1245 |
+
"grad_norm": 0.011489451862871647,
|
| 1246 |
+
"learning_rate": 6.61764705882353e-06,
|
| 1247 |
+
"loss": 0.0273,
|
| 1248 |
+
"step": 88500
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 2.6176470588235294,
|
| 1252 |
+
"grad_norm": 0.0028730102349072695,
|
| 1253 |
+
"learning_rate": 6.372549019607843e-06,
|
| 1254 |
+
"loss": 0.0277,
|
| 1255 |
+
"step": 89000
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 2.6323529411764706,
|
| 1259 |
+
"grad_norm": 0.005050596781075001,
|
| 1260 |
+
"learning_rate": 6.127450980392157e-06,
|
| 1261 |
+
"loss": 0.0148,
|
| 1262 |
+
"step": 89500
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 2.6470588235294117,
|
| 1266 |
+
"grad_norm": 0.0028206182178109884,
|
| 1267 |
+
"learning_rate": 5.882352941176471e-06,
|
| 1268 |
+
"loss": 0.0121,
|
| 1269 |
+
"step": 90000
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 2.661764705882353,
|
| 1273 |
+
"grad_norm": 0.0025763895828276873,
|
| 1274 |
+
"learning_rate": 5.637254901960784e-06,
|
| 1275 |
+
"loss": 0.0119,
|
| 1276 |
+
"step": 90500
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 2.6764705882352944,
|
| 1280 |
+
"grad_norm": 0.024479951709508896,
|
| 1281 |
+
"learning_rate": 5.392156862745099e-06,
|
| 1282 |
+
"loss": 0.0141,
|
| 1283 |
+
"step": 91000
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 2.6911764705882355,
|
| 1287 |
+
"grad_norm": 0.001322206575423479,
|
| 1288 |
+
"learning_rate": 5.147058823529412e-06,
|
| 1289 |
+
"loss": 0.0104,
|
| 1290 |
+
"step": 91500
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 2.7058823529411766,
|
| 1294 |
+
"grad_norm": 0.0011514847865328193,
|
| 1295 |
+
"learning_rate": 4.901960784313726e-06,
|
| 1296 |
+
"loss": 0.0192,
|
| 1297 |
+
"step": 92000
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 2.7205882352941178,
|
| 1301 |
+
"grad_norm": 0.002985934726893902,
|
| 1302 |
+
"learning_rate": 4.65686274509804e-06,
|
| 1303 |
+
"loss": 0.018,
|
| 1304 |
+
"step": 92500
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 2.735294117647059,
|
| 1308 |
+
"grad_norm": 0.0024135473649948835,
|
| 1309 |
+
"learning_rate": 4.411764705882353e-06,
|
| 1310 |
+
"loss": 0.009,
|
| 1311 |
+
"step": 93000
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 2.75,
|
| 1315 |
+
"grad_norm": 0.001011635409668088,
|
| 1316 |
+
"learning_rate": 4.166666666666667e-06,
|
| 1317 |
+
"loss": 0.014,
|
| 1318 |
+
"step": 93500
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 2.764705882352941,
|
| 1322 |
+
"grad_norm": 1.9080002307891846,
|
| 1323 |
+
"learning_rate": 3.92156862745098e-06,
|
| 1324 |
+
"loss": 0.0191,
|
| 1325 |
+
"step": 94000
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 2.7794117647058822,
|
| 1329 |
+
"grad_norm": 0.004457990638911724,
|
| 1330 |
+
"learning_rate": 3.6764705882352942e-06,
|
| 1331 |
+
"loss": 0.0058,
|
| 1332 |
+
"step": 94500
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 2.7941176470588234,
|
| 1336 |
+
"grad_norm": 0.0034340699203312397,
|
| 1337 |
+
"learning_rate": 3.431372549019608e-06,
|
| 1338 |
+
"loss": 0.0124,
|
| 1339 |
+
"step": 95000
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 2.8088235294117645,
|
| 1343 |
+
"grad_norm": 0.0019747125916182995,
|
| 1344 |
+
"learning_rate": 3.1862745098039216e-06,
|
| 1345 |
+
"loss": 0.0244,
|
| 1346 |
+
"step": 95500
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 2.8235294117647056,
|
| 1350 |
+
"grad_norm": 0.0038515429478138685,
|
| 1351 |
+
"learning_rate": 2.9411764705882355e-06,
|
| 1352 |
+
"loss": 0.0163,
|
| 1353 |
+
"step": 96000
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 2.838235294117647,
|
| 1357 |
+
"grad_norm": 0.0029344563372433186,
|
| 1358 |
+
"learning_rate": 2.6960784313725493e-06,
|
| 1359 |
+
"loss": 0.0098,
|
| 1360 |
+
"step": 96500
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 2.8529411764705883,
|
| 1364 |
+
"grad_norm": 0.005202002823352814,
|
| 1365 |
+
"learning_rate": 2.450980392156863e-06,
|
| 1366 |
+
"loss": 0.0098,
|
| 1367 |
+
"step": 97000
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 2.8676470588235294,
|
| 1371 |
+
"grad_norm": 0.0045198979787528515,
|
| 1372 |
+
"learning_rate": 2.2058823529411767e-06,
|
| 1373 |
+
"loss": 0.0129,
|
| 1374 |
+
"step": 97500
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 2.8823529411764706,
|
| 1378 |
+
"grad_norm": 0.002258594846352935,
|
| 1379 |
+
"learning_rate": 1.96078431372549e-06,
|
| 1380 |
+
"loss": 0.0063,
|
| 1381 |
+
"step": 98000
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 2.8970588235294117,
|
| 1385 |
+
"grad_norm": 0.036095574498176575,
|
| 1386 |
+
"learning_rate": 1.715686274509804e-06,
|
| 1387 |
+
"loss": 0.0153,
|
| 1388 |
+
"step": 98500
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 2.911764705882353,
|
| 1392 |
+
"grad_norm": 0.003401304828003049,
|
| 1393 |
+
"learning_rate": 1.4705882352941177e-06,
|
| 1394 |
+
"loss": 0.0098,
|
| 1395 |
+
"step": 99000
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 2.9264705882352944,
|
| 1399 |
+
"grad_norm": 0.004960035905241966,
|
| 1400 |
+
"learning_rate": 1.2254901960784314e-06,
|
| 1401 |
+
"loss": 0.0072,
|
| 1402 |
+
"step": 99500
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 2.9411764705882355,
|
| 1406 |
+
"grad_norm": 0.004606081638485193,
|
| 1407 |
+
"learning_rate": 9.80392156862745e-07,
|
| 1408 |
+
"loss": 0.0107,
|
| 1409 |
+
"step": 100000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 2.9558823529411766,
|
| 1413 |
+
"grad_norm": 0.0029280243907123804,
|
| 1414 |
+
"learning_rate": 7.352941176470589e-07,
|
| 1415 |
+
"loss": 0.0078,
|
| 1416 |
+
"step": 100500
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 2.9705882352941178,
|
| 1420 |
+
"grad_norm": 0.016250332817435265,
|
| 1421 |
+
"learning_rate": 4.901960784313725e-07,
|
| 1422 |
+
"loss": 0.0183,
|
| 1423 |
+
"step": 101000
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 2.985294117647059,
|
| 1427 |
+
"grad_norm": 0.002401071134954691,
|
| 1428 |
+
"learning_rate": 2.4509803921568627e-07,
|
| 1429 |
+
"loss": 0.0117,
|
| 1430 |
+
"step": 101500
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 3.0,
|
| 1434 |
+
"grad_norm": 0.0015339870005846024,
|
| 1435 |
+
"learning_rate": 0.0,
|
| 1436 |
+
"loss": 0.0134,
|
| 1437 |
+
"step": 102000
|
| 1438 |
+
}
|
| 1439 |
+
],
|
| 1440 |
+
"logging_steps": 500,
|
| 1441 |
+
"max_steps": 102000,
|
| 1442 |
+
"num_input_tokens_seen": 0,
|
| 1443 |
+
"num_train_epochs": 3,
|
| 1444 |
+
"save_steps": 500,
|
| 1445 |
+
"total_flos": 6.323969768236646e+19,
|
| 1446 |
+
"train_batch_size": 8,
|
| 1447 |
+
"trial_name": null,
|
| 1448 |
+
"trial_params": null
|
| 1449 |
+
}
|
checkpoint-102000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cd4d39b2c10c5bbc511dbb6611ec40b6fde59c4b12ef3b86a49c36eefffee464
|
| 3 |
+
size 4984
|
config.json
ADDED
|
@@ -0,0 +1,55 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google/vit-base-patch16-224-in21k",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"ViTForImageClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"encoder_stride": 16,
|
| 8 |
+
"finetuning_task": "image-classification",
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.0,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "bb",
|
| 14 |
+
"1": "bk",
|
| 15 |
+
"10": "wp",
|
| 16 |
+
"11": "wq",
|
| 17 |
+
"12": "wr",
|
| 18 |
+
"2": "bn",
|
| 19 |
+
"3": "bp",
|
| 20 |
+
"4": "bq",
|
| 21 |
+
"5": "br",
|
| 22 |
+
"6": "empty",
|
| 23 |
+
"7": "wb",
|
| 24 |
+
"8": "wk",
|
| 25 |
+
"9": "wn"
|
| 26 |
+
},
|
| 27 |
+
"image_size": 224,
|
| 28 |
+
"initializer_range": 0.02,
|
| 29 |
+
"intermediate_size": 3072,
|
| 30 |
+
"label2id": {
|
| 31 |
+
"bb": "0",
|
| 32 |
+
"bk": "1",
|
| 33 |
+
"bn": "2",
|
| 34 |
+
"bp": "3",
|
| 35 |
+
"bq": "4",
|
| 36 |
+
"br": "5",
|
| 37 |
+
"empty": "6",
|
| 38 |
+
"wb": "7",
|
| 39 |
+
"wk": "8",
|
| 40 |
+
"wn": "9",
|
| 41 |
+
"wp": "10",
|
| 42 |
+
"wq": "11",
|
| 43 |
+
"wr": "12"
|
| 44 |
+
},
|
| 45 |
+
"layer_norm_eps": 1e-12,
|
| 46 |
+
"model_type": "vit",
|
| 47 |
+
"num_attention_heads": 12,
|
| 48 |
+
"num_channels": 3,
|
| 49 |
+
"num_hidden_layers": 12,
|
| 50 |
+
"patch_size": 16,
|
| 51 |
+
"problem_type": "single_label_classification",
|
| 52 |
+
"qkv_bias": true,
|
| 53 |
+
"torch_dtype": "float32",
|
| 54 |
+
"transformers_version": "4.40.2"
|
| 55 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0fed5d30904017d84562df1d3f165e9382aa1a8c48564aa75d45eed1462f356e
|
| 3 |
+
size 343257812
|
preprocessor_config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_valid_processor_keys": [
|
| 3 |
+
"images",
|
| 4 |
+
"do_resize",
|
| 5 |
+
"size",
|
| 6 |
+
"resample",
|
| 7 |
+
"do_rescale",
|
| 8 |
+
"rescale_factor",
|
| 9 |
+
"do_normalize",
|
| 10 |
+
"image_mean",
|
| 11 |
+
"image_std",
|
| 12 |
+
"return_tensors",
|
| 13 |
+
"data_format",
|
| 14 |
+
"input_data_format"
|
| 15 |
+
],
|
| 16 |
+
"do_normalize": true,
|
| 17 |
+
"do_rescale": true,
|
| 18 |
+
"do_resize": true,
|
| 19 |
+
"image_mean": [
|
| 20 |
+
0.5,
|
| 21 |
+
0.5,
|
| 22 |
+
0.5
|
| 23 |
+
],
|
| 24 |
+
"image_processor_type": "ViTImageProcessor",
|
| 25 |
+
"image_std": [
|
| 26 |
+
0.5,
|
| 27 |
+
0.5,
|
| 28 |
+
0.5
|
| 29 |
+
],
|
| 30 |
+
"resample": 2,
|
| 31 |
+
"rescale_factor": 0.00392156862745098,
|
| 32 |
+
"size": {
|
| 33 |
+
"height": 224,
|
| 34 |
+
"width": 224
|
| 35 |
+
}
|
| 36 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 3.0,
|
| 3 |
+
"total_flos": 6.323969768236646e+19,
|
| 4 |
+
"train_loss": 0.0325347986571929,
|
| 5 |
+
"train_runtime": 10422.0944,
|
| 6 |
+
"train_samples_per_second": 78.295,
|
| 7 |
+
"train_steps_per_second": 9.787
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1458 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 3.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 102000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.014705882352941176,
|
| 13 |
+
"grad_norm": 0.35731062293052673,
|
| 14 |
+
"learning_rate": 4.975490196078432e-05,
|
| 15 |
+
"loss": 0.5354,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.029411764705882353,
|
| 20 |
+
"grad_norm": 2.1595866680145264,
|
| 21 |
+
"learning_rate": 4.9509803921568634e-05,
|
| 22 |
+
"loss": 0.1627,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.04411764705882353,
|
| 27 |
+
"grad_norm": 0.14825384318828583,
|
| 28 |
+
"learning_rate": 4.9264705882352944e-05,
|
| 29 |
+
"loss": 0.101,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.058823529411764705,
|
| 34 |
+
"grad_norm": 0.07121703773736954,
|
| 35 |
+
"learning_rate": 4.901960784313725e-05,
|
| 36 |
+
"loss": 0.0952,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.07352941176470588,
|
| 41 |
+
"grad_norm": 0.06089532747864723,
|
| 42 |
+
"learning_rate": 4.877450980392157e-05,
|
| 43 |
+
"loss": 0.0881,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.08823529411764706,
|
| 48 |
+
"grad_norm": 0.037034619599580765,
|
| 49 |
+
"learning_rate": 4.8529411764705885e-05,
|
| 50 |
+
"loss": 0.0717,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.10294117647058823,
|
| 55 |
+
"grad_norm": 0.01999847963452339,
|
| 56 |
+
"learning_rate": 4.82843137254902e-05,
|
| 57 |
+
"loss": 0.0848,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.11764705882352941,
|
| 62 |
+
"grad_norm": 0.018650399520993233,
|
| 63 |
+
"learning_rate": 4.803921568627452e-05,
|
| 64 |
+
"loss": 0.068,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.1323529411764706,
|
| 69 |
+
"grad_norm": 0.023023229092359543,
|
| 70 |
+
"learning_rate": 4.7794117647058826e-05,
|
| 71 |
+
"loss": 0.0626,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.14705882352941177,
|
| 76 |
+
"grad_norm": 0.018715515732765198,
|
| 77 |
+
"learning_rate": 4.7549019607843135e-05,
|
| 78 |
+
"loss": 0.0555,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.16176470588235295,
|
| 83 |
+
"grad_norm": 0.11842140555381775,
|
| 84 |
+
"learning_rate": 4.730392156862745e-05,
|
| 85 |
+
"loss": 0.0658,
|
| 86 |
+
"step": 5500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.17647058823529413,
|
| 90 |
+
"grad_norm": 0.04816881939768791,
|
| 91 |
+
"learning_rate": 4.705882352941177e-05,
|
| 92 |
+
"loss": 0.0648,
|
| 93 |
+
"step": 6000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 0.19117647058823528,
|
| 97 |
+
"grad_norm": 0.025240018963813782,
|
| 98 |
+
"learning_rate": 4.681372549019608e-05,
|
| 99 |
+
"loss": 0.0632,
|
| 100 |
+
"step": 6500
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 0.20588235294117646,
|
| 104 |
+
"grad_norm": 0.007390766404569149,
|
| 105 |
+
"learning_rate": 4.656862745098039e-05,
|
| 106 |
+
"loss": 0.0705,
|
| 107 |
+
"step": 7000
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 0.22058823529411764,
|
| 111 |
+
"grad_norm": 0.011664963327348232,
|
| 112 |
+
"learning_rate": 4.632352941176471e-05,
|
| 113 |
+
"loss": 0.0552,
|
| 114 |
+
"step": 7500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 0.23529411764705882,
|
| 118 |
+
"grad_norm": 39.69175720214844,
|
| 119 |
+
"learning_rate": 4.607843137254902e-05,
|
| 120 |
+
"loss": 0.0686,
|
| 121 |
+
"step": 8000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 0.25,
|
| 125 |
+
"grad_norm": 0.007366931065917015,
|
| 126 |
+
"learning_rate": 4.5833333333333334e-05,
|
| 127 |
+
"loss": 0.061,
|
| 128 |
+
"step": 8500
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.2647058823529412,
|
| 132 |
+
"grad_norm": 0.003396671498194337,
|
| 133 |
+
"learning_rate": 4.558823529411765e-05,
|
| 134 |
+
"loss": 0.0464,
|
| 135 |
+
"step": 9000
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 0.27941176470588236,
|
| 139 |
+
"grad_norm": 0.045366521924734116,
|
| 140 |
+
"learning_rate": 4.5343137254901966e-05,
|
| 141 |
+
"loss": 0.053,
|
| 142 |
+
"step": 9500
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 0.29411764705882354,
|
| 146 |
+
"grad_norm": 1.137495756149292,
|
| 147 |
+
"learning_rate": 4.5098039215686275e-05,
|
| 148 |
+
"loss": 0.0484,
|
| 149 |
+
"step": 10000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.3088235294117647,
|
| 153 |
+
"grad_norm": 0.0062417215667665005,
|
| 154 |
+
"learning_rate": 4.485294117647059e-05,
|
| 155 |
+
"loss": 0.0564,
|
| 156 |
+
"step": 10500
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 0.3235294117647059,
|
| 160 |
+
"grad_norm": 0.003990447614341974,
|
| 161 |
+
"learning_rate": 4.460784313725491e-05,
|
| 162 |
+
"loss": 0.0364,
|
| 163 |
+
"step": 11000
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 0.3382352941176471,
|
| 167 |
+
"grad_norm": 0.11569799482822418,
|
| 168 |
+
"learning_rate": 4.4362745098039216e-05,
|
| 169 |
+
"loss": 0.0436,
|
| 170 |
+
"step": 11500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.35294117647058826,
|
| 174 |
+
"grad_norm": 0.01159907691180706,
|
| 175 |
+
"learning_rate": 4.411764705882353e-05,
|
| 176 |
+
"loss": 0.0456,
|
| 177 |
+
"step": 12000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 0.36764705882352944,
|
| 181 |
+
"grad_norm": 10.926911354064941,
|
| 182 |
+
"learning_rate": 4.387254901960784e-05,
|
| 183 |
+
"loss": 0.0464,
|
| 184 |
+
"step": 12500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 0.38235294117647056,
|
| 188 |
+
"grad_norm": 0.03444543853402138,
|
| 189 |
+
"learning_rate": 4.362745098039216e-05,
|
| 190 |
+
"loss": 0.0512,
|
| 191 |
+
"step": 13000
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 0.39705882352941174,
|
| 195 |
+
"grad_norm": 0.002395658055320382,
|
| 196 |
+
"learning_rate": 4.3382352941176474e-05,
|
| 197 |
+
"loss": 0.0586,
|
| 198 |
+
"step": 13500
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 0.4117647058823529,
|
| 202 |
+
"grad_norm": 0.010224021971225739,
|
| 203 |
+
"learning_rate": 4.313725490196079e-05,
|
| 204 |
+
"loss": 0.0407,
|
| 205 |
+
"step": 14000
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 0.4264705882352941,
|
| 209 |
+
"grad_norm": 0.010642035864293575,
|
| 210 |
+
"learning_rate": 4.28921568627451e-05,
|
| 211 |
+
"loss": 0.0536,
|
| 212 |
+
"step": 14500
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.4411764705882353,
|
| 216 |
+
"grad_norm": 0.00970557238906622,
|
| 217 |
+
"learning_rate": 4.2647058823529415e-05,
|
| 218 |
+
"loss": 0.0526,
|
| 219 |
+
"step": 15000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 0.45588235294117646,
|
| 223 |
+
"grad_norm": 0.0030253385193645954,
|
| 224 |
+
"learning_rate": 4.2401960784313724e-05,
|
| 225 |
+
"loss": 0.0432,
|
| 226 |
+
"step": 15500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 0.47058823529411764,
|
| 230 |
+
"grad_norm": 0.5423064827919006,
|
| 231 |
+
"learning_rate": 4.215686274509804e-05,
|
| 232 |
+
"loss": 0.0381,
|
| 233 |
+
"step": 16000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 0.4852941176470588,
|
| 237 |
+
"grad_norm": 7.454125881195068,
|
| 238 |
+
"learning_rate": 4.1911764705882356e-05,
|
| 239 |
+
"loss": 0.0397,
|
| 240 |
+
"step": 16500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 0.5,
|
| 244 |
+
"grad_norm": 3.87744402885437,
|
| 245 |
+
"learning_rate": 4.166666666666667e-05,
|
| 246 |
+
"loss": 0.0564,
|
| 247 |
+
"step": 17000
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 0.5147058823529411,
|
| 251 |
+
"grad_norm": 33.39067840576172,
|
| 252 |
+
"learning_rate": 4.142156862745099e-05,
|
| 253 |
+
"loss": 0.0455,
|
| 254 |
+
"step": 17500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.5294117647058824,
|
| 258 |
+
"grad_norm": 0.008246080949902534,
|
| 259 |
+
"learning_rate": 4.11764705882353e-05,
|
| 260 |
+
"loss": 0.0515,
|
| 261 |
+
"step": 18000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 0.5441176470588235,
|
| 265 |
+
"grad_norm": 0.018734918907284737,
|
| 266 |
+
"learning_rate": 4.0931372549019607e-05,
|
| 267 |
+
"loss": 0.0453,
|
| 268 |
+
"step": 18500
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 0.5588235294117647,
|
| 272 |
+
"grad_norm": 0.007373027969151735,
|
| 273 |
+
"learning_rate": 4.068627450980392e-05,
|
| 274 |
+
"loss": 0.0485,
|
| 275 |
+
"step": 19000
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 0.5735294117647058,
|
| 279 |
+
"grad_norm": 0.0023267469368875027,
|
| 280 |
+
"learning_rate": 4.044117647058824e-05,
|
| 281 |
+
"loss": 0.0372,
|
| 282 |
+
"step": 19500
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 0.5882352941176471,
|
| 286 |
+
"grad_norm": 0.1311252862215042,
|
| 287 |
+
"learning_rate": 4.0196078431372555e-05,
|
| 288 |
+
"loss": 0.0442,
|
| 289 |
+
"step": 20000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 0.6029411764705882,
|
| 293 |
+
"grad_norm": 0.003710985416546464,
|
| 294 |
+
"learning_rate": 3.9950980392156864e-05,
|
| 295 |
+
"loss": 0.0585,
|
| 296 |
+
"step": 20500
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.6176470588235294,
|
| 300 |
+
"grad_norm": 0.004861747846007347,
|
| 301 |
+
"learning_rate": 3.970588235294117e-05,
|
| 302 |
+
"loss": 0.0374,
|
| 303 |
+
"step": 21000
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 0.6323529411764706,
|
| 307 |
+
"grad_norm": 0.004351571202278137,
|
| 308 |
+
"learning_rate": 3.946078431372549e-05,
|
| 309 |
+
"loss": 0.0265,
|
| 310 |
+
"step": 21500
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 0.6470588235294118,
|
| 314 |
+
"grad_norm": 0.008851751685142517,
|
| 315 |
+
"learning_rate": 3.9215686274509805e-05,
|
| 316 |
+
"loss": 0.0314,
|
| 317 |
+
"step": 22000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 0.6617647058823529,
|
| 321 |
+
"grad_norm": 0.0046307104639709,
|
| 322 |
+
"learning_rate": 3.897058823529412e-05,
|
| 323 |
+
"loss": 0.0445,
|
| 324 |
+
"step": 22500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 0.6764705882352942,
|
| 328 |
+
"grad_norm": 0.002735880669206381,
|
| 329 |
+
"learning_rate": 3.872549019607844e-05,
|
| 330 |
+
"loss": 0.0433,
|
| 331 |
+
"step": 23000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 0.6911764705882353,
|
| 335 |
+
"grad_norm": 0.0218490082770586,
|
| 336 |
+
"learning_rate": 3.8480392156862746e-05,
|
| 337 |
+
"loss": 0.0444,
|
| 338 |
+
"step": 23500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.7058823529411765,
|
| 342 |
+
"grad_norm": 2.371448278427124,
|
| 343 |
+
"learning_rate": 3.8235294117647055e-05,
|
| 344 |
+
"loss": 0.033,
|
| 345 |
+
"step": 24000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 0.7205882352941176,
|
| 349 |
+
"grad_norm": 0.005015780217945576,
|
| 350 |
+
"learning_rate": 3.799019607843137e-05,
|
| 351 |
+
"loss": 0.0422,
|
| 352 |
+
"step": 24500
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 0.7352941176470589,
|
| 356 |
+
"grad_norm": 1.3102850914001465,
|
| 357 |
+
"learning_rate": 3.774509803921569e-05,
|
| 358 |
+
"loss": 0.035,
|
| 359 |
+
"step": 25000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 0.75,
|
| 363 |
+
"grad_norm": 0.004522955510765314,
|
| 364 |
+
"learning_rate": 3.7500000000000003e-05,
|
| 365 |
+
"loss": 0.0303,
|
| 366 |
+
"step": 25500
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 0.7647058823529411,
|
| 370 |
+
"grad_norm": 0.006077844649553299,
|
| 371 |
+
"learning_rate": 3.725490196078432e-05,
|
| 372 |
+
"loss": 0.0503,
|
| 373 |
+
"step": 26000
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 0.7794117647058824,
|
| 377 |
+
"grad_norm": 0.0026446939446032047,
|
| 378 |
+
"learning_rate": 3.700980392156863e-05,
|
| 379 |
+
"loss": 0.0264,
|
| 380 |
+
"step": 26500
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.7941176470588235,
|
| 384 |
+
"grad_norm": 0.004485867917537689,
|
| 385 |
+
"learning_rate": 3.6764705882352945e-05,
|
| 386 |
+
"loss": 0.0388,
|
| 387 |
+
"step": 27000
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 0.8088235294117647,
|
| 391 |
+
"grad_norm": 0.0073866695165634155,
|
| 392 |
+
"learning_rate": 3.6519607843137254e-05,
|
| 393 |
+
"loss": 0.0283,
|
| 394 |
+
"step": 27500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 0.8235294117647058,
|
| 398 |
+
"grad_norm": 0.012984287925064564,
|
| 399 |
+
"learning_rate": 3.627450980392157e-05,
|
| 400 |
+
"loss": 0.0393,
|
| 401 |
+
"step": 28000
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 0.8382352941176471,
|
| 405 |
+
"grad_norm": 0.01751883700489998,
|
| 406 |
+
"learning_rate": 3.6029411764705886e-05,
|
| 407 |
+
"loss": 0.0388,
|
| 408 |
+
"step": 28500
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 0.8529411764705882,
|
| 412 |
+
"grad_norm": 0.0035843336954712868,
|
| 413 |
+
"learning_rate": 3.5784313725490195e-05,
|
| 414 |
+
"loss": 0.0384,
|
| 415 |
+
"step": 29000
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 0.8676470588235294,
|
| 419 |
+
"grad_norm": 0.0645672082901001,
|
| 420 |
+
"learning_rate": 3.553921568627451e-05,
|
| 421 |
+
"loss": 0.0378,
|
| 422 |
+
"step": 29500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.8823529411764706,
|
| 426 |
+
"grad_norm": 0.008566264994442463,
|
| 427 |
+
"learning_rate": 3.529411764705883e-05,
|
| 428 |
+
"loss": 0.0315,
|
| 429 |
+
"step": 30000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 0.8970588235294118,
|
| 433 |
+
"grad_norm": 0.010571740567684174,
|
| 434 |
+
"learning_rate": 3.5049019607843136e-05,
|
| 435 |
+
"loss": 0.0324,
|
| 436 |
+
"step": 30500
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 0.9117647058823529,
|
| 440 |
+
"grad_norm": 0.0022533361334353685,
|
| 441 |
+
"learning_rate": 3.480392156862745e-05,
|
| 442 |
+
"loss": 0.0292,
|
| 443 |
+
"step": 31000
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 0.9264705882352942,
|
| 447 |
+
"grad_norm": 0.006164130289107561,
|
| 448 |
+
"learning_rate": 3.455882352941177e-05,
|
| 449 |
+
"loss": 0.0407,
|
| 450 |
+
"step": 31500
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 0.9411764705882353,
|
| 454 |
+
"grad_norm": 0.007435985840857029,
|
| 455 |
+
"learning_rate": 3.431372549019608e-05,
|
| 456 |
+
"loss": 0.0449,
|
| 457 |
+
"step": 32000
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 0.9558823529411765,
|
| 461 |
+
"grad_norm": 0.003777585458010435,
|
| 462 |
+
"learning_rate": 3.4068627450980394e-05,
|
| 463 |
+
"loss": 0.0247,
|
| 464 |
+
"step": 32500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.9705882352941176,
|
| 468 |
+
"grad_norm": 0.005975374951958656,
|
| 469 |
+
"learning_rate": 3.382352941176471e-05,
|
| 470 |
+
"loss": 0.0309,
|
| 471 |
+
"step": 33000
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 0.9852941176470589,
|
| 475 |
+
"grad_norm": 0.0016012012492865324,
|
| 476 |
+
"learning_rate": 3.357843137254902e-05,
|
| 477 |
+
"loss": 0.0294,
|
| 478 |
+
"step": 33500
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 1.0,
|
| 482 |
+
"grad_norm": 0.020027656108140945,
|
| 483 |
+
"learning_rate": 3.3333333333333335e-05,
|
| 484 |
+
"loss": 0.022,
|
| 485 |
+
"step": 34000
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 1.0147058823529411,
|
| 489 |
+
"grad_norm": 10.916418075561523,
|
| 490 |
+
"learning_rate": 3.308823529411765e-05,
|
| 491 |
+
"loss": 0.0264,
|
| 492 |
+
"step": 34500
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 1.0294117647058822,
|
| 496 |
+
"grad_norm": 0.013728056102991104,
|
| 497 |
+
"learning_rate": 3.284313725490196e-05,
|
| 498 |
+
"loss": 0.032,
|
| 499 |
+
"step": 35000
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 1.0441176470588236,
|
| 503 |
+
"grad_norm": 0.02546643279492855,
|
| 504 |
+
"learning_rate": 3.2598039215686276e-05,
|
| 505 |
+
"loss": 0.0404,
|
| 506 |
+
"step": 35500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 1.0588235294117647,
|
| 510 |
+
"grad_norm": 0.004185052588582039,
|
| 511 |
+
"learning_rate": 3.235294117647059e-05,
|
| 512 |
+
"loss": 0.0306,
|
| 513 |
+
"step": 36000
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 1.0735294117647058,
|
| 517 |
+
"grad_norm": 0.016653403639793396,
|
| 518 |
+
"learning_rate": 3.210784313725491e-05,
|
| 519 |
+
"loss": 0.0217,
|
| 520 |
+
"step": 36500
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 1.088235294117647,
|
| 524 |
+
"grad_norm": 0.0028331661596894264,
|
| 525 |
+
"learning_rate": 3.186274509803922e-05,
|
| 526 |
+
"loss": 0.0265,
|
| 527 |
+
"step": 37000
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 1.1029411764705883,
|
| 531 |
+
"grad_norm": 0.013931985944509506,
|
| 532 |
+
"learning_rate": 3.161764705882353e-05,
|
| 533 |
+
"loss": 0.0399,
|
| 534 |
+
"step": 37500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 1.1176470588235294,
|
| 538 |
+
"grad_norm": 0.001668413169682026,
|
| 539 |
+
"learning_rate": 3.137254901960784e-05,
|
| 540 |
+
"loss": 0.0202,
|
| 541 |
+
"step": 38000
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 1.1323529411764706,
|
| 545 |
+
"grad_norm": 0.00567347789183259,
|
| 546 |
+
"learning_rate": 3.112745098039216e-05,
|
| 547 |
+
"loss": 0.0336,
|
| 548 |
+
"step": 38500
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 1.1470588235294117,
|
| 552 |
+
"grad_norm": 0.003477458842098713,
|
| 553 |
+
"learning_rate": 3.0882352941176475e-05,
|
| 554 |
+
"loss": 0.0253,
|
| 555 |
+
"step": 39000
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 1.161764705882353,
|
| 559 |
+
"grad_norm": 0.21819466352462769,
|
| 560 |
+
"learning_rate": 3.063725490196079e-05,
|
| 561 |
+
"loss": 0.0293,
|
| 562 |
+
"step": 39500
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 1.1764705882352942,
|
| 566 |
+
"grad_norm": 0.004708552733063698,
|
| 567 |
+
"learning_rate": 3.0392156862745097e-05,
|
| 568 |
+
"loss": 0.0268,
|
| 569 |
+
"step": 40000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 1.1911764705882353,
|
| 573 |
+
"grad_norm": 0.0018363581039011478,
|
| 574 |
+
"learning_rate": 3.0147058823529413e-05,
|
| 575 |
+
"loss": 0.0266,
|
| 576 |
+
"step": 40500
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 1.2058823529411764,
|
| 580 |
+
"grad_norm": 0.005581580102443695,
|
| 581 |
+
"learning_rate": 2.9901960784313725e-05,
|
| 582 |
+
"loss": 0.0361,
|
| 583 |
+
"step": 41000
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 1.2205882352941178,
|
| 587 |
+
"grad_norm": 0.0035322746261954308,
|
| 588 |
+
"learning_rate": 2.965686274509804e-05,
|
| 589 |
+
"loss": 0.0274,
|
| 590 |
+
"step": 41500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 1.2352941176470589,
|
| 594 |
+
"grad_norm": 0.0038708180654793978,
|
| 595 |
+
"learning_rate": 2.9411764705882354e-05,
|
| 596 |
+
"loss": 0.0241,
|
| 597 |
+
"step": 42000
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 1.25,
|
| 601 |
+
"grad_norm": 0.0019606975838541985,
|
| 602 |
+
"learning_rate": 2.916666666666667e-05,
|
| 603 |
+
"loss": 0.0219,
|
| 604 |
+
"step": 42500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 1.2647058823529411,
|
| 608 |
+
"grad_norm": 0.00241417670622468,
|
| 609 |
+
"learning_rate": 2.8921568627450986e-05,
|
| 610 |
+
"loss": 0.0244,
|
| 611 |
+
"step": 43000
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 1.2794117647058822,
|
| 615 |
+
"grad_norm": 0.003593308152630925,
|
| 616 |
+
"learning_rate": 2.8676470588235295e-05,
|
| 617 |
+
"loss": 0.027,
|
| 618 |
+
"step": 43500
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 1.2941176470588236,
|
| 622 |
+
"grad_norm": 1.4003372192382812,
|
| 623 |
+
"learning_rate": 2.8431372549019608e-05,
|
| 624 |
+
"loss": 0.0392,
|
| 625 |
+
"step": 44000
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 1.3088235294117647,
|
| 629 |
+
"grad_norm": 0.41921645402908325,
|
| 630 |
+
"learning_rate": 2.8186274509803924e-05,
|
| 631 |
+
"loss": 0.0299,
|
| 632 |
+
"step": 44500
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 1.3235294117647058,
|
| 636 |
+
"grad_norm": 1.567896842956543,
|
| 637 |
+
"learning_rate": 2.7941176470588236e-05,
|
| 638 |
+
"loss": 0.0282,
|
| 639 |
+
"step": 45000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 1.3382352941176472,
|
| 643 |
+
"grad_norm": 0.003143745008856058,
|
| 644 |
+
"learning_rate": 2.7696078431372552e-05,
|
| 645 |
+
"loss": 0.0197,
|
| 646 |
+
"step": 45500
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 1.3529411764705883,
|
| 650 |
+
"grad_norm": 0.0016428233357146382,
|
| 651 |
+
"learning_rate": 2.7450980392156865e-05,
|
| 652 |
+
"loss": 0.0344,
|
| 653 |
+
"step": 46000
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 1.3676470588235294,
|
| 657 |
+
"grad_norm": 0.005276167765259743,
|
| 658 |
+
"learning_rate": 2.7205882352941174e-05,
|
| 659 |
+
"loss": 0.0312,
|
| 660 |
+
"step": 46500
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 1.3823529411764706,
|
| 664 |
+
"grad_norm": 0.004461635369807482,
|
| 665 |
+
"learning_rate": 2.696078431372549e-05,
|
| 666 |
+
"loss": 0.025,
|
| 667 |
+
"step": 47000
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 1.3970588235294117,
|
| 671 |
+
"grad_norm": 0.003238542238250375,
|
| 672 |
+
"learning_rate": 2.6715686274509806e-05,
|
| 673 |
+
"loss": 0.0236,
|
| 674 |
+
"step": 47500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 1.4117647058823528,
|
| 678 |
+
"grad_norm": 0.0017602238804101944,
|
| 679 |
+
"learning_rate": 2.647058823529412e-05,
|
| 680 |
+
"loss": 0.0234,
|
| 681 |
+
"step": 48000
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 1.4264705882352942,
|
| 685 |
+
"grad_norm": 0.002768098609521985,
|
| 686 |
+
"learning_rate": 2.6225490196078435e-05,
|
| 687 |
+
"loss": 0.026,
|
| 688 |
+
"step": 48500
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 1.4411764705882353,
|
| 692 |
+
"grad_norm": 0.002579926745966077,
|
| 693 |
+
"learning_rate": 2.5980392156862747e-05,
|
| 694 |
+
"loss": 0.0167,
|
| 695 |
+
"step": 49000
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 1.4558823529411764,
|
| 699 |
+
"grad_norm": 0.008888750337064266,
|
| 700 |
+
"learning_rate": 2.5735294117647057e-05,
|
| 701 |
+
"loss": 0.0288,
|
| 702 |
+
"step": 49500
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 1.4705882352941178,
|
| 706 |
+
"grad_norm": 0.0038515792693942785,
|
| 707 |
+
"learning_rate": 2.5490196078431373e-05,
|
| 708 |
+
"loss": 0.0252,
|
| 709 |
+
"step": 50000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 1.4852941176470589,
|
| 713 |
+
"grad_norm": 4.428643226623535,
|
| 714 |
+
"learning_rate": 2.5245098039215685e-05,
|
| 715 |
+
"loss": 0.016,
|
| 716 |
+
"step": 50500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 1.5,
|
| 720 |
+
"grad_norm": 0.002629584399983287,
|
| 721 |
+
"learning_rate": 2.5e-05,
|
| 722 |
+
"loss": 0.0214,
|
| 723 |
+
"step": 51000
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 1.5147058823529411,
|
| 727 |
+
"grad_norm": 0.004097859375178814,
|
| 728 |
+
"learning_rate": 2.4754901960784317e-05,
|
| 729 |
+
"loss": 0.0183,
|
| 730 |
+
"step": 51500
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 1.5294117647058822,
|
| 734 |
+
"grad_norm": 0.00568019924685359,
|
| 735 |
+
"learning_rate": 2.4509803921568626e-05,
|
| 736 |
+
"loss": 0.0271,
|
| 737 |
+
"step": 52000
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 1.5441176470588234,
|
| 741 |
+
"grad_norm": 0.0020534582436084747,
|
| 742 |
+
"learning_rate": 2.4264705882352942e-05,
|
| 743 |
+
"loss": 0.0281,
|
| 744 |
+
"step": 52500
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 1.5588235294117647,
|
| 748 |
+
"grad_norm": 0.003300599753856659,
|
| 749 |
+
"learning_rate": 2.401960784313726e-05,
|
| 750 |
+
"loss": 0.0284,
|
| 751 |
+
"step": 53000
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 1.5735294117647058,
|
| 755 |
+
"grad_norm": 0.001472037984058261,
|
| 756 |
+
"learning_rate": 2.3774509803921568e-05,
|
| 757 |
+
"loss": 0.0205,
|
| 758 |
+
"step": 53500
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 1.5882352941176472,
|
| 762 |
+
"grad_norm": 0.002835978288203478,
|
| 763 |
+
"learning_rate": 2.3529411764705884e-05,
|
| 764 |
+
"loss": 0.0234,
|
| 765 |
+
"step": 54000
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 1.6029411764705883,
|
| 769 |
+
"grad_norm": 0.003979724366217852,
|
| 770 |
+
"learning_rate": 2.3284313725490196e-05,
|
| 771 |
+
"loss": 0.0233,
|
| 772 |
+
"step": 54500
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 1.6176470588235294,
|
| 776 |
+
"grad_norm": 0.0036057273391634226,
|
| 777 |
+
"learning_rate": 2.303921568627451e-05,
|
| 778 |
+
"loss": 0.0319,
|
| 779 |
+
"step": 55000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 1.6323529411764706,
|
| 783 |
+
"grad_norm": 0.0024822901468724012,
|
| 784 |
+
"learning_rate": 2.2794117647058825e-05,
|
| 785 |
+
"loss": 0.0211,
|
| 786 |
+
"step": 55500
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 1.6470588235294117,
|
| 790 |
+
"grad_norm": 0.008930359967052937,
|
| 791 |
+
"learning_rate": 2.2549019607843138e-05,
|
| 792 |
+
"loss": 0.0309,
|
| 793 |
+
"step": 56000
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 1.6617647058823528,
|
| 797 |
+
"grad_norm": 1.3381928205490112,
|
| 798 |
+
"learning_rate": 2.2303921568627454e-05,
|
| 799 |
+
"loss": 0.019,
|
| 800 |
+
"step": 56500
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 1.6764705882352942,
|
| 804 |
+
"grad_norm": 0.004481327719986439,
|
| 805 |
+
"learning_rate": 2.2058823529411766e-05,
|
| 806 |
+
"loss": 0.0228,
|
| 807 |
+
"step": 57000
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 1.6911764705882353,
|
| 811 |
+
"grad_norm": 0.012597435154020786,
|
| 812 |
+
"learning_rate": 2.181372549019608e-05,
|
| 813 |
+
"loss": 0.0285,
|
| 814 |
+
"step": 57500
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 1.7058823529411766,
|
| 818 |
+
"grad_norm": 0.0034783107694238424,
|
| 819 |
+
"learning_rate": 2.1568627450980395e-05,
|
| 820 |
+
"loss": 0.0165,
|
| 821 |
+
"step": 58000
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 1.7205882352941178,
|
| 825 |
+
"grad_norm": 0.0032083301339298487,
|
| 826 |
+
"learning_rate": 2.1323529411764707e-05,
|
| 827 |
+
"loss": 0.0306,
|
| 828 |
+
"step": 58500
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 1.7352941176470589,
|
| 832 |
+
"grad_norm": 0.0039000194519758224,
|
| 833 |
+
"learning_rate": 2.107843137254902e-05,
|
| 834 |
+
"loss": 0.0199,
|
| 835 |
+
"step": 59000
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 1.75,
|
| 839 |
+
"grad_norm": 0.0034919639583677053,
|
| 840 |
+
"learning_rate": 2.0833333333333336e-05,
|
| 841 |
+
"loss": 0.034,
|
| 842 |
+
"step": 59500
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 1.7647058823529411,
|
| 846 |
+
"grad_norm": 0.004928025882691145,
|
| 847 |
+
"learning_rate": 2.058823529411765e-05,
|
| 848 |
+
"loss": 0.0213,
|
| 849 |
+
"step": 60000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 1.7794117647058822,
|
| 853 |
+
"grad_norm": 0.003211489412933588,
|
| 854 |
+
"learning_rate": 2.034313725490196e-05,
|
| 855 |
+
"loss": 0.0206,
|
| 856 |
+
"step": 60500
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 1.7941176470588234,
|
| 860 |
+
"grad_norm": 0.011239697225391865,
|
| 861 |
+
"learning_rate": 2.0098039215686277e-05,
|
| 862 |
+
"loss": 0.0201,
|
| 863 |
+
"step": 61000
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 1.8088235294117647,
|
| 867 |
+
"grad_norm": 0.0024609589017927647,
|
| 868 |
+
"learning_rate": 1.9852941176470586e-05,
|
| 869 |
+
"loss": 0.0286,
|
| 870 |
+
"step": 61500
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 1.8235294117647058,
|
| 874 |
+
"grad_norm": 0.0046806493774056435,
|
| 875 |
+
"learning_rate": 1.9607843137254903e-05,
|
| 876 |
+
"loss": 0.0216,
|
| 877 |
+
"step": 62000
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 1.8382352941176472,
|
| 881 |
+
"grad_norm": 2.30717396736145,
|
| 882 |
+
"learning_rate": 1.936274509803922e-05,
|
| 883 |
+
"loss": 0.0246,
|
| 884 |
+
"step": 62500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 1.8529411764705883,
|
| 888 |
+
"grad_norm": 0.00428669573739171,
|
| 889 |
+
"learning_rate": 1.9117647058823528e-05,
|
| 890 |
+
"loss": 0.0166,
|
| 891 |
+
"step": 63000
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 1.8676470588235294,
|
| 895 |
+
"grad_norm": 0.011403021402657032,
|
| 896 |
+
"learning_rate": 1.8872549019607844e-05,
|
| 897 |
+
"loss": 0.0272,
|
| 898 |
+
"step": 63500
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 1.8823529411764706,
|
| 902 |
+
"grad_norm": 0.0065813250839710236,
|
| 903 |
+
"learning_rate": 1.862745098039216e-05,
|
| 904 |
+
"loss": 0.0217,
|
| 905 |
+
"step": 64000
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 1.8970588235294117,
|
| 909 |
+
"grad_norm": 0.0024323465768247843,
|
| 910 |
+
"learning_rate": 1.8382352941176472e-05,
|
| 911 |
+
"loss": 0.0176,
|
| 912 |
+
"step": 64500
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 1.9117647058823528,
|
| 916 |
+
"grad_norm": 0.001616469700820744,
|
| 917 |
+
"learning_rate": 1.8137254901960785e-05,
|
| 918 |
+
"loss": 0.0169,
|
| 919 |
+
"step": 65000
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 1.9264705882352942,
|
| 923 |
+
"grad_norm": 0.004322522785514593,
|
| 924 |
+
"learning_rate": 1.7892156862745098e-05,
|
| 925 |
+
"loss": 0.0216,
|
| 926 |
+
"step": 65500
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 1.9411764705882353,
|
| 930 |
+
"grad_norm": 0.0024695000611245632,
|
| 931 |
+
"learning_rate": 1.7647058823529414e-05,
|
| 932 |
+
"loss": 0.032,
|
| 933 |
+
"step": 66000
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 1.9558823529411766,
|
| 937 |
+
"grad_norm": 0.010675052180886269,
|
| 938 |
+
"learning_rate": 1.7401960784313726e-05,
|
| 939 |
+
"loss": 0.023,
|
| 940 |
+
"step": 66500
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 1.9705882352941178,
|
| 944 |
+
"grad_norm": 0.49467232823371887,
|
| 945 |
+
"learning_rate": 1.715686274509804e-05,
|
| 946 |
+
"loss": 0.0169,
|
| 947 |
+
"step": 67000
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 1.9852941176470589,
|
| 951 |
+
"grad_norm": 0.01075649168342352,
|
| 952 |
+
"learning_rate": 1.6911764705882355e-05,
|
| 953 |
+
"loss": 0.0177,
|
| 954 |
+
"step": 67500
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 2.0,
|
| 958 |
+
"grad_norm": 0.026703685522079468,
|
| 959 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 960 |
+
"loss": 0.0259,
|
| 961 |
+
"step": 68000
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 2.014705882352941,
|
| 965 |
+
"grad_norm": 0.0023890878073871136,
|
| 966 |
+
"learning_rate": 1.642156862745098e-05,
|
| 967 |
+
"loss": 0.0128,
|
| 968 |
+
"step": 68500
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 2.0294117647058822,
|
| 972 |
+
"grad_norm": 0.0015898487763479352,
|
| 973 |
+
"learning_rate": 1.6176470588235296e-05,
|
| 974 |
+
"loss": 0.0112,
|
| 975 |
+
"step": 69000
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 2.0441176470588234,
|
| 979 |
+
"grad_norm": 0.0041648312471807,
|
| 980 |
+
"learning_rate": 1.593137254901961e-05,
|
| 981 |
+
"loss": 0.0155,
|
| 982 |
+
"step": 69500
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 2.0588235294117645,
|
| 986 |
+
"grad_norm": 0.10951696336269379,
|
| 987 |
+
"learning_rate": 1.568627450980392e-05,
|
| 988 |
+
"loss": 0.0225,
|
| 989 |
+
"step": 70000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 2.073529411764706,
|
| 993 |
+
"grad_norm": 0.0021414640359580517,
|
| 994 |
+
"learning_rate": 1.5441176470588237e-05,
|
| 995 |
+
"loss": 0.0174,
|
| 996 |
+
"step": 70500
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 2.088235294117647,
|
| 1000 |
+
"grad_norm": 0.001627126126550138,
|
| 1001 |
+
"learning_rate": 1.5196078431372548e-05,
|
| 1002 |
+
"loss": 0.0154,
|
| 1003 |
+
"step": 71000
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 2.1029411764705883,
|
| 1007 |
+
"grad_norm": 0.005821748171001673,
|
| 1008 |
+
"learning_rate": 1.4950980392156863e-05,
|
| 1009 |
+
"loss": 0.0255,
|
| 1010 |
+
"step": 71500
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 2.1176470588235294,
|
| 1014 |
+
"grad_norm": 0.0071876379661262035,
|
| 1015 |
+
"learning_rate": 1.4705882352941177e-05,
|
| 1016 |
+
"loss": 0.024,
|
| 1017 |
+
"step": 72000
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 2.1323529411764706,
|
| 1021 |
+
"grad_norm": 0.00294076488353312,
|
| 1022 |
+
"learning_rate": 1.4460784313725493e-05,
|
| 1023 |
+
"loss": 0.012,
|
| 1024 |
+
"step": 72500
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 2.1470588235294117,
|
| 1028 |
+
"grad_norm": 0.004260234069079161,
|
| 1029 |
+
"learning_rate": 1.4215686274509804e-05,
|
| 1030 |
+
"loss": 0.0202,
|
| 1031 |
+
"step": 73000
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 2.161764705882353,
|
| 1035 |
+
"grad_norm": 0.005722737871110439,
|
| 1036 |
+
"learning_rate": 1.3970588235294118e-05,
|
| 1037 |
+
"loss": 0.0283,
|
| 1038 |
+
"step": 73500
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 2.176470588235294,
|
| 1042 |
+
"grad_norm": 0.016162721440196037,
|
| 1043 |
+
"learning_rate": 1.3725490196078432e-05,
|
| 1044 |
+
"loss": 0.0115,
|
| 1045 |
+
"step": 74000
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 2.1911764705882355,
|
| 1049 |
+
"grad_norm": 0.005048350431025028,
|
| 1050 |
+
"learning_rate": 1.3480392156862745e-05,
|
| 1051 |
+
"loss": 0.023,
|
| 1052 |
+
"step": 74500
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 2.2058823529411766,
|
| 1056 |
+
"grad_norm": 0.004102902952581644,
|
| 1057 |
+
"learning_rate": 1.323529411764706e-05,
|
| 1058 |
+
"loss": 0.018,
|
| 1059 |
+
"step": 75000
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 2.2205882352941178,
|
| 1063 |
+
"grad_norm": 0.0021536860149353743,
|
| 1064 |
+
"learning_rate": 1.2990196078431374e-05,
|
| 1065 |
+
"loss": 0.0169,
|
| 1066 |
+
"step": 75500
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 2.235294117647059,
|
| 1070 |
+
"grad_norm": 0.001363063813187182,
|
| 1071 |
+
"learning_rate": 1.2745098039215686e-05,
|
| 1072 |
+
"loss": 0.0129,
|
| 1073 |
+
"step": 76000
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 2.25,
|
| 1077 |
+
"grad_norm": 0.0025944672524929047,
|
| 1078 |
+
"learning_rate": 1.25e-05,
|
| 1079 |
+
"loss": 0.0095,
|
| 1080 |
+
"step": 76500
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 2.264705882352941,
|
| 1084 |
+
"grad_norm": 0.001098868204280734,
|
| 1085 |
+
"learning_rate": 1.2254901960784313e-05,
|
| 1086 |
+
"loss": 0.0183,
|
| 1087 |
+
"step": 77000
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 2.2794117647058822,
|
| 1091 |
+
"grad_norm": 0.00421817135065794,
|
| 1092 |
+
"learning_rate": 1.200980392156863e-05,
|
| 1093 |
+
"loss": 0.0214,
|
| 1094 |
+
"step": 77500
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 2.2941176470588234,
|
| 1098 |
+
"grad_norm": 0.0016445108922198415,
|
| 1099 |
+
"learning_rate": 1.1764705882352942e-05,
|
| 1100 |
+
"loss": 0.009,
|
| 1101 |
+
"step": 78000
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 2.3088235294117645,
|
| 1105 |
+
"grad_norm": 0.002265740418806672,
|
| 1106 |
+
"learning_rate": 1.1519607843137254e-05,
|
| 1107 |
+
"loss": 0.0161,
|
| 1108 |
+
"step": 78500
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 2.323529411764706,
|
| 1112 |
+
"grad_norm": 0.007099386304616928,
|
| 1113 |
+
"learning_rate": 1.1274509803921569e-05,
|
| 1114 |
+
"loss": 0.0143,
|
| 1115 |
+
"step": 79000
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 2.338235294117647,
|
| 1119 |
+
"grad_norm": 0.003914414439350367,
|
| 1120 |
+
"learning_rate": 1.1029411764705883e-05,
|
| 1121 |
+
"loss": 0.0115,
|
| 1122 |
+
"step": 79500
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 2.3529411764705883,
|
| 1126 |
+
"grad_norm": 0.002513893647119403,
|
| 1127 |
+
"learning_rate": 1.0784313725490197e-05,
|
| 1128 |
+
"loss": 0.0136,
|
| 1129 |
+
"step": 80000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 2.3676470588235294,
|
| 1133 |
+
"grad_norm": 0.0027356306090950966,
|
| 1134 |
+
"learning_rate": 1.053921568627451e-05,
|
| 1135 |
+
"loss": 0.0247,
|
| 1136 |
+
"step": 80500
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 2.3823529411764706,
|
| 1140 |
+
"grad_norm": 0.0817839726805687,
|
| 1141 |
+
"learning_rate": 1.0294117647058824e-05,
|
| 1142 |
+
"loss": 0.0169,
|
| 1143 |
+
"step": 81000
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 2.3970588235294117,
|
| 1147 |
+
"grad_norm": 0.08163878321647644,
|
| 1148 |
+
"learning_rate": 1.0049019607843139e-05,
|
| 1149 |
+
"loss": 0.0114,
|
| 1150 |
+
"step": 81500
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 2.411764705882353,
|
| 1154 |
+
"grad_norm": 0.002625884721055627,
|
| 1155 |
+
"learning_rate": 9.803921568627451e-06,
|
| 1156 |
+
"loss": 0.0147,
|
| 1157 |
+
"step": 82000
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 2.426470588235294,
|
| 1161 |
+
"grad_norm": 0.004531237296760082,
|
| 1162 |
+
"learning_rate": 9.558823529411764e-06,
|
| 1163 |
+
"loss": 0.0106,
|
| 1164 |
+
"step": 82500
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 2.4411764705882355,
|
| 1168 |
+
"grad_norm": 0.0032483581453561783,
|
| 1169 |
+
"learning_rate": 9.31372549019608e-06,
|
| 1170 |
+
"loss": 0.0183,
|
| 1171 |
+
"step": 83000
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 2.4558823529411766,
|
| 1175 |
+
"grad_norm": 0.0050782193429768085,
|
| 1176 |
+
"learning_rate": 9.068627450980392e-06,
|
| 1177 |
+
"loss": 0.0168,
|
| 1178 |
+
"step": 83500
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 2.4705882352941178,
|
| 1182 |
+
"grad_norm": 0.003839900717139244,
|
| 1183 |
+
"learning_rate": 8.823529411764707e-06,
|
| 1184 |
+
"loss": 0.0209,
|
| 1185 |
+
"step": 84000
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 2.485294117647059,
|
| 1189 |
+
"grad_norm": 0.008910595439374447,
|
| 1190 |
+
"learning_rate": 8.57843137254902e-06,
|
| 1191 |
+
"loss": 0.0271,
|
| 1192 |
+
"step": 84500
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 2.5,
|
| 1196 |
+
"grad_norm": 0.008313077501952648,
|
| 1197 |
+
"learning_rate": 8.333333333333334e-06,
|
| 1198 |
+
"loss": 0.0175,
|
| 1199 |
+
"step": 85000
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 2.514705882352941,
|
| 1203 |
+
"grad_norm": 0.0029984668362885714,
|
| 1204 |
+
"learning_rate": 8.088235294117648e-06,
|
| 1205 |
+
"loss": 0.0084,
|
| 1206 |
+
"step": 85500
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 2.5294117647058822,
|
| 1210 |
+
"grad_norm": 0.004524989053606987,
|
| 1211 |
+
"learning_rate": 7.84313725490196e-06,
|
| 1212 |
+
"loss": 0.0156,
|
| 1213 |
+
"step": 86000
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 2.5441176470588234,
|
| 1217 |
+
"grad_norm": 0.0023315059952437878,
|
| 1218 |
+
"learning_rate": 7.598039215686274e-06,
|
| 1219 |
+
"loss": 0.0121,
|
| 1220 |
+
"step": 86500
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 2.5588235294117645,
|
| 1224 |
+
"grad_norm": 0.0010077670449391007,
|
| 1225 |
+
"learning_rate": 7.3529411764705884e-06,
|
| 1226 |
+
"loss": 0.0113,
|
| 1227 |
+
"step": 87000
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 2.5735294117647056,
|
| 1231 |
+
"grad_norm": 0.0012684785760939121,
|
| 1232 |
+
"learning_rate": 7.107843137254902e-06,
|
| 1233 |
+
"loss": 0.0204,
|
| 1234 |
+
"step": 87500
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 2.588235294117647,
|
| 1238 |
+
"grad_norm": 0.004918810911476612,
|
| 1239 |
+
"learning_rate": 6.862745098039216e-06,
|
| 1240 |
+
"loss": 0.0059,
|
| 1241 |
+
"step": 88000
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 2.6029411764705883,
|
| 1245 |
+
"grad_norm": 0.011489451862871647,
|
| 1246 |
+
"learning_rate": 6.61764705882353e-06,
|
| 1247 |
+
"loss": 0.0273,
|
| 1248 |
+
"step": 88500
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 2.6176470588235294,
|
| 1252 |
+
"grad_norm": 0.0028730102349072695,
|
| 1253 |
+
"learning_rate": 6.372549019607843e-06,
|
| 1254 |
+
"loss": 0.0277,
|
| 1255 |
+
"step": 89000
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 2.6323529411764706,
|
| 1259 |
+
"grad_norm": 0.005050596781075001,
|
| 1260 |
+
"learning_rate": 6.127450980392157e-06,
|
| 1261 |
+
"loss": 0.0148,
|
| 1262 |
+
"step": 89500
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 2.6470588235294117,
|
| 1266 |
+
"grad_norm": 0.0028206182178109884,
|
| 1267 |
+
"learning_rate": 5.882352941176471e-06,
|
| 1268 |
+
"loss": 0.0121,
|
| 1269 |
+
"step": 90000
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 2.661764705882353,
|
| 1273 |
+
"grad_norm": 0.0025763895828276873,
|
| 1274 |
+
"learning_rate": 5.637254901960784e-06,
|
| 1275 |
+
"loss": 0.0119,
|
| 1276 |
+
"step": 90500
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 2.6764705882352944,
|
| 1280 |
+
"grad_norm": 0.024479951709508896,
|
| 1281 |
+
"learning_rate": 5.392156862745099e-06,
|
| 1282 |
+
"loss": 0.0141,
|
| 1283 |
+
"step": 91000
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 2.6911764705882355,
|
| 1287 |
+
"grad_norm": 0.001322206575423479,
|
| 1288 |
+
"learning_rate": 5.147058823529412e-06,
|
| 1289 |
+
"loss": 0.0104,
|
| 1290 |
+
"step": 91500
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 2.7058823529411766,
|
| 1294 |
+
"grad_norm": 0.0011514847865328193,
|
| 1295 |
+
"learning_rate": 4.901960784313726e-06,
|
| 1296 |
+
"loss": 0.0192,
|
| 1297 |
+
"step": 92000
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 2.7205882352941178,
|
| 1301 |
+
"grad_norm": 0.002985934726893902,
|
| 1302 |
+
"learning_rate": 4.65686274509804e-06,
|
| 1303 |
+
"loss": 0.018,
|
| 1304 |
+
"step": 92500
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 2.735294117647059,
|
| 1308 |
+
"grad_norm": 0.0024135473649948835,
|
| 1309 |
+
"learning_rate": 4.411764705882353e-06,
|
| 1310 |
+
"loss": 0.009,
|
| 1311 |
+
"step": 93000
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 2.75,
|
| 1315 |
+
"grad_norm": 0.001011635409668088,
|
| 1316 |
+
"learning_rate": 4.166666666666667e-06,
|
| 1317 |
+
"loss": 0.014,
|
| 1318 |
+
"step": 93500
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 2.764705882352941,
|
| 1322 |
+
"grad_norm": 1.9080002307891846,
|
| 1323 |
+
"learning_rate": 3.92156862745098e-06,
|
| 1324 |
+
"loss": 0.0191,
|
| 1325 |
+
"step": 94000
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 2.7794117647058822,
|
| 1329 |
+
"grad_norm": 0.004457990638911724,
|
| 1330 |
+
"learning_rate": 3.6764705882352942e-06,
|
| 1331 |
+
"loss": 0.0058,
|
| 1332 |
+
"step": 94500
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 2.7941176470588234,
|
| 1336 |
+
"grad_norm": 0.0034340699203312397,
|
| 1337 |
+
"learning_rate": 3.431372549019608e-06,
|
| 1338 |
+
"loss": 0.0124,
|
| 1339 |
+
"step": 95000
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 2.8088235294117645,
|
| 1343 |
+
"grad_norm": 0.0019747125916182995,
|
| 1344 |
+
"learning_rate": 3.1862745098039216e-06,
|
| 1345 |
+
"loss": 0.0244,
|
| 1346 |
+
"step": 95500
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 2.8235294117647056,
|
| 1350 |
+
"grad_norm": 0.0038515429478138685,
|
| 1351 |
+
"learning_rate": 2.9411764705882355e-06,
|
| 1352 |
+
"loss": 0.0163,
|
| 1353 |
+
"step": 96000
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 2.838235294117647,
|
| 1357 |
+
"grad_norm": 0.0029344563372433186,
|
| 1358 |
+
"learning_rate": 2.6960784313725493e-06,
|
| 1359 |
+
"loss": 0.0098,
|
| 1360 |
+
"step": 96500
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 2.8529411764705883,
|
| 1364 |
+
"grad_norm": 0.005202002823352814,
|
| 1365 |
+
"learning_rate": 2.450980392156863e-06,
|
| 1366 |
+
"loss": 0.0098,
|
| 1367 |
+
"step": 97000
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 2.8676470588235294,
|
| 1371 |
+
"grad_norm": 0.0045198979787528515,
|
| 1372 |
+
"learning_rate": 2.2058823529411767e-06,
|
| 1373 |
+
"loss": 0.0129,
|
| 1374 |
+
"step": 97500
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 2.8823529411764706,
|
| 1378 |
+
"grad_norm": 0.002258594846352935,
|
| 1379 |
+
"learning_rate": 1.96078431372549e-06,
|
| 1380 |
+
"loss": 0.0063,
|
| 1381 |
+
"step": 98000
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 2.8970588235294117,
|
| 1385 |
+
"grad_norm": 0.036095574498176575,
|
| 1386 |
+
"learning_rate": 1.715686274509804e-06,
|
| 1387 |
+
"loss": 0.0153,
|
| 1388 |
+
"step": 98500
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 2.911764705882353,
|
| 1392 |
+
"grad_norm": 0.003401304828003049,
|
| 1393 |
+
"learning_rate": 1.4705882352941177e-06,
|
| 1394 |
+
"loss": 0.0098,
|
| 1395 |
+
"step": 99000
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 2.9264705882352944,
|
| 1399 |
+
"grad_norm": 0.004960035905241966,
|
| 1400 |
+
"learning_rate": 1.2254901960784314e-06,
|
| 1401 |
+
"loss": 0.0072,
|
| 1402 |
+
"step": 99500
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 2.9411764705882355,
|
| 1406 |
+
"grad_norm": 0.004606081638485193,
|
| 1407 |
+
"learning_rate": 9.80392156862745e-07,
|
| 1408 |
+
"loss": 0.0107,
|
| 1409 |
+
"step": 100000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 2.9558823529411766,
|
| 1413 |
+
"grad_norm": 0.0029280243907123804,
|
| 1414 |
+
"learning_rate": 7.352941176470589e-07,
|
| 1415 |
+
"loss": 0.0078,
|
| 1416 |
+
"step": 100500
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 2.9705882352941178,
|
| 1420 |
+
"grad_norm": 0.016250332817435265,
|
| 1421 |
+
"learning_rate": 4.901960784313725e-07,
|
| 1422 |
+
"loss": 0.0183,
|
| 1423 |
+
"step": 101000
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 2.985294117647059,
|
| 1427 |
+
"grad_norm": 0.002401071134954691,
|
| 1428 |
+
"learning_rate": 2.4509803921568627e-07,
|
| 1429 |
+
"loss": 0.0117,
|
| 1430 |
+
"step": 101500
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 3.0,
|
| 1434 |
+
"grad_norm": 0.0015339870005846024,
|
| 1435 |
+
"learning_rate": 0.0,
|
| 1436 |
+
"loss": 0.0134,
|
| 1437 |
+
"step": 102000
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 3.0,
|
| 1441 |
+
"step": 102000,
|
| 1442 |
+
"total_flos": 6.323969768236646e+19,
|
| 1443 |
+
"train_loss": 0.0325347986571929,
|
| 1444 |
+
"train_runtime": 10422.0944,
|
| 1445 |
+
"train_samples_per_second": 78.295,
|
| 1446 |
+
"train_steps_per_second": 9.787
|
| 1447 |
+
}
|
| 1448 |
+
],
|
| 1449 |
+
"logging_steps": 500,
|
| 1450 |
+
"max_steps": 102000,
|
| 1451 |
+
"num_input_tokens_seen": 0,
|
| 1452 |
+
"num_train_epochs": 3,
|
| 1453 |
+
"save_steps": 500,
|
| 1454 |
+
"total_flos": 6.323969768236646e+19,
|
| 1455 |
+
"train_batch_size": 8,
|
| 1456 |
+
"trial_name": null,
|
| 1457 |
+
"trial_params": null
|
| 1458 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:cd4d39b2c10c5bbc511dbb6611ec40b6fde59c4b12ef3b86a49c36eefffee464
|
| 3 |
+
size 4984
|