Upload 15 files
Browse files- README.md +57 -0
- added_tokens.json +9 -0
- all_results.json +16 -0
- config.json +39 -0
- eval_results.json +10 -0
- generation_config.json +6 -0
- merges.txt +0 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +30 -0
- tokenizer.json +0 -0
- tokenizer_config.json +76 -0
- train_results.json +9 -0
- trainer_state.json +2079 -0
- training_args.bin +3 -0
- vocab.json +0 -0
README.md
ADDED
|
@@ -0,0 +1,57 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: nferruz/ProtGPT2
|
| 4 |
+
tags:
|
| 5 |
+
- generated_from_trainer
|
| 6 |
+
metrics:
|
| 7 |
+
- accuracy
|
| 8 |
+
model-index:
|
| 9 |
+
- name: model_output_31_2
|
| 10 |
+
results: []
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
<!-- This model card has been generated automatically according to the information the Trainer had access to. You
|
| 14 |
+
should probably proofread and complete it, then remove this comment. -->
|
| 15 |
+
|
| 16 |
+
# model_output_31_2
|
| 17 |
+
|
| 18 |
+
This model is a fine-tuned version of [nferruz/ProtGPT2](https://huggingface.co/nferruz/ProtGPT2) on an unknown dataset.
|
| 19 |
+
It achieves the following results on the evaluation set:
|
| 20 |
+
- Loss: 6.4472
|
| 21 |
+
- Accuracy: 0.6444
|
| 22 |
+
|
| 23 |
+
## Model description
|
| 24 |
+
|
| 25 |
+
More information needed
|
| 26 |
+
|
| 27 |
+
## Intended uses & limitations
|
| 28 |
+
|
| 29 |
+
More information needed
|
| 30 |
+
|
| 31 |
+
## Training and evaluation data
|
| 32 |
+
|
| 33 |
+
More information needed
|
| 34 |
+
|
| 35 |
+
## Training procedure
|
| 36 |
+
|
| 37 |
+
### Training hyperparameters
|
| 38 |
+
|
| 39 |
+
The following hyperparameters were used during training:
|
| 40 |
+
- learning_rate: 0.001
|
| 41 |
+
- train_batch_size: 8
|
| 42 |
+
- eval_batch_size: 8
|
| 43 |
+
- seed: 42
|
| 44 |
+
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
|
| 45 |
+
- lr_scheduler_type: linear
|
| 46 |
+
- num_epochs: 200.0
|
| 47 |
+
|
| 48 |
+
### Training results
|
| 49 |
+
|
| 50 |
+
|
| 51 |
+
|
| 52 |
+
### Framework versions
|
| 53 |
+
|
| 54 |
+
- Transformers 4.42.0.dev0
|
| 55 |
+
- Pytorch 2.0.1
|
| 56 |
+
- Datasets 2.19.1
|
| 57 |
+
- Tokenizers 0.19.1
|
added_tokens.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<PAD>": 50263,
|
| 3 |
+
"<endoftext>": 50262,
|
| 4 |
+
"<startoftext>": 50261,
|
| 5 |
+
"LABEL:": 50258,
|
| 6 |
+
"NEGATIVE": 50260,
|
| 7 |
+
"POSITIVE": 50259,
|
| 8 |
+
"SEQUENCE:": 50257
|
| 9 |
+
}
|
all_results.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 200.0,
|
| 3 |
+
"eval_accuracy": 0.6444462546613084,
|
| 4 |
+
"eval_loss": 6.447216033935547,
|
| 5 |
+
"eval_runtime": 43.5897,
|
| 6 |
+
"eval_samples": 324,
|
| 7 |
+
"eval_samples_per_second": 7.433,
|
| 8 |
+
"eval_steps_per_second": 0.941,
|
| 9 |
+
"perplexity": 630.9433206955741,
|
| 10 |
+
"total_flos": 2.53654993993728e+18,
|
| 11 |
+
"train_loss": 0.006510871606680919,
|
| 12 |
+
"train_runtime": 172601.0106,
|
| 13 |
+
"train_samples": 2914,
|
| 14 |
+
"train_samples_per_second": 3.377,
|
| 15 |
+
"train_steps_per_second": 0.845
|
| 16 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,39 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "nferruz/ProtGPT2",
|
| 3 |
+
"activation_function": "gelu_new",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"GPT2LMHeadModel"
|
| 6 |
+
],
|
| 7 |
+
"attn_pdrop": 0.1,
|
| 8 |
+
"bos_token_id": 0,
|
| 9 |
+
"embd_pdrop": 0.1,
|
| 10 |
+
"eos_token_id": 0,
|
| 11 |
+
"initializer_range": 0.02,
|
| 12 |
+
"layer_norm_epsilon": 1e-05,
|
| 13 |
+
"model_type": "gpt2",
|
| 14 |
+
"n_ctx": 1024,
|
| 15 |
+
"n_embd": 1280,
|
| 16 |
+
"n_head": 20,
|
| 17 |
+
"n_inner": null,
|
| 18 |
+
"n_layer": 36,
|
| 19 |
+
"n_positions": 1024,
|
| 20 |
+
"reorder_and_upcast_attn": false,
|
| 21 |
+
"resid_pdrop": 0.1,
|
| 22 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 23 |
+
"scale_attn_weights": true,
|
| 24 |
+
"summary_activation": null,
|
| 25 |
+
"summary_first_dropout": 0.1,
|
| 26 |
+
"summary_proj_to_labels": true,
|
| 27 |
+
"summary_type": "cls_index",
|
| 28 |
+
"summary_use_proj": true,
|
| 29 |
+
"task_specific_params": {
|
| 30 |
+
"text-generation": {
|
| 31 |
+
"do_sample": true,
|
| 32 |
+
"max_length": 50
|
| 33 |
+
}
|
| 34 |
+
},
|
| 35 |
+
"torch_dtype": "float32",
|
| 36 |
+
"transformers_version": "4.42.0.dev0",
|
| 37 |
+
"use_cache": true,
|
| 38 |
+
"vocab_size": 50264
|
| 39 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 200.0,
|
| 3 |
+
"eval_accuracy": 0.6444462546613084,
|
| 4 |
+
"eval_loss": 6.447216033935547,
|
| 5 |
+
"eval_runtime": 43.5897,
|
| 6 |
+
"eval_samples": 324,
|
| 7 |
+
"eval_samples_per_second": 7.433,
|
| 8 |
+
"eval_steps_per_second": 0.941,
|
| 9 |
+
"perplexity": 630.9433206955741
|
| 10 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"eos_token_id": 0,
|
| 5 |
+
"transformers_version": "4.42.0.dev0"
|
| 6 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:307fd963c96172fb246971ac766f40bc97b5e91e4209836af58da69128a43794
|
| 3 |
+
size 3096298525
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<startoftext>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": false,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<endoftext>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": false,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"pad_token": {
|
| 17 |
+
"content": "<PAD>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": false,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
},
|
| 23 |
+
"unk_token": {
|
| 24 |
+
"content": "<|endoftext|>",
|
| 25 |
+
"lstrip": false,
|
| 26 |
+
"normalized": true,
|
| 27 |
+
"rstrip": false,
|
| 28 |
+
"single_word": false
|
| 29 |
+
}
|
| 30 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,76 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"added_tokens_decoder": {
|
| 4 |
+
"0": {
|
| 5 |
+
"content": "<|endoftext|>",
|
| 6 |
+
"lstrip": false,
|
| 7 |
+
"normalized": true,
|
| 8 |
+
"rstrip": false,
|
| 9 |
+
"single_word": false,
|
| 10 |
+
"special": true
|
| 11 |
+
},
|
| 12 |
+
"50257": {
|
| 13 |
+
"content": "SEQUENCE:",
|
| 14 |
+
"lstrip": false,
|
| 15 |
+
"normalized": true,
|
| 16 |
+
"rstrip": false,
|
| 17 |
+
"single_word": false,
|
| 18 |
+
"special": false
|
| 19 |
+
},
|
| 20 |
+
"50258": {
|
| 21 |
+
"content": "LABEL:",
|
| 22 |
+
"lstrip": false,
|
| 23 |
+
"normalized": true,
|
| 24 |
+
"rstrip": false,
|
| 25 |
+
"single_word": false,
|
| 26 |
+
"special": false
|
| 27 |
+
},
|
| 28 |
+
"50259": {
|
| 29 |
+
"content": "POSITIVE",
|
| 30 |
+
"lstrip": false,
|
| 31 |
+
"normalized": true,
|
| 32 |
+
"rstrip": false,
|
| 33 |
+
"single_word": false,
|
| 34 |
+
"special": false
|
| 35 |
+
},
|
| 36 |
+
"50260": {
|
| 37 |
+
"content": "NEGATIVE",
|
| 38 |
+
"lstrip": false,
|
| 39 |
+
"normalized": true,
|
| 40 |
+
"rstrip": false,
|
| 41 |
+
"single_word": false,
|
| 42 |
+
"special": false
|
| 43 |
+
},
|
| 44 |
+
"50261": {
|
| 45 |
+
"content": "<startoftext>",
|
| 46 |
+
"lstrip": false,
|
| 47 |
+
"normalized": false,
|
| 48 |
+
"rstrip": false,
|
| 49 |
+
"single_word": false,
|
| 50 |
+
"special": true
|
| 51 |
+
},
|
| 52 |
+
"50262": {
|
| 53 |
+
"content": "<endoftext>",
|
| 54 |
+
"lstrip": false,
|
| 55 |
+
"normalized": false,
|
| 56 |
+
"rstrip": false,
|
| 57 |
+
"single_word": false,
|
| 58 |
+
"special": true
|
| 59 |
+
},
|
| 60 |
+
"50263": {
|
| 61 |
+
"content": "<PAD>",
|
| 62 |
+
"lstrip": false,
|
| 63 |
+
"normalized": false,
|
| 64 |
+
"rstrip": false,
|
| 65 |
+
"single_word": false,
|
| 66 |
+
"special": true
|
| 67 |
+
}
|
| 68 |
+
},
|
| 69 |
+
"bos_token": "<startoftext>",
|
| 70 |
+
"clean_up_tokenization_spaces": true,
|
| 71 |
+
"eos_token": "<endoftext>",
|
| 72 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 73 |
+
"pad_token": "<PAD>",
|
| 74 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 75 |
+
"unk_token": "<|endoftext|>"
|
| 76 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 200.0,
|
| 3 |
+
"total_flos": 2.53654993993728e+18,
|
| 4 |
+
"train_loss": 0.006510871606680919,
|
| 5 |
+
"train_runtime": 172601.0106,
|
| 6 |
+
"train_samples": 2914,
|
| 7 |
+
"train_samples_per_second": 3.377,
|
| 8 |
+
"train_steps_per_second": 0.845
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,2079 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 200.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 145800,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.6858710562414266,
|
| 13 |
+
"grad_norm": 1.2456575632095337,
|
| 14 |
+
"learning_rate": 0.000996570644718793,
|
| 15 |
+
"loss": 2.6328,
|
| 16 |
+
"step": 500
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 1.3717421124828533,
|
| 20 |
+
"grad_norm": 0.6728461384773254,
|
| 21 |
+
"learning_rate": 0.0009931412894375858,
|
| 22 |
+
"loss": 1.9106,
|
| 23 |
+
"step": 1000
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 2.05761316872428,
|
| 27 |
+
"grad_norm": 0.7172486782073975,
|
| 28 |
+
"learning_rate": 0.0009897119341563787,
|
| 29 |
+
"loss": 1.5385,
|
| 30 |
+
"step": 1500
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 2.7434842249657065,
|
| 34 |
+
"grad_norm": 0.5451925992965698,
|
| 35 |
+
"learning_rate": 0.0009862825788751715,
|
| 36 |
+
"loss": 1.1861,
|
| 37 |
+
"step": 2000
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 3.4293552812071333,
|
| 41 |
+
"grad_norm": 1.0264705419540405,
|
| 42 |
+
"learning_rate": 0.0009828532235939644,
|
| 43 |
+
"loss": 0.9558,
|
| 44 |
+
"step": 2500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 4.11522633744856,
|
| 48 |
+
"grad_norm": 0.5801027417182922,
|
| 49 |
+
"learning_rate": 0.0009794238683127573,
|
| 50 |
+
"loss": 0.7722,
|
| 51 |
+
"step": 3000
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 4.801097393689986,
|
| 55 |
+
"grad_norm": 0.6234860420227051,
|
| 56 |
+
"learning_rate": 0.0009759945130315501,
|
| 57 |
+
"loss": 0.5497,
|
| 58 |
+
"step": 3500
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 5.486968449931413,
|
| 62 |
+
"grad_norm": 0.4896549880504608,
|
| 63 |
+
"learning_rate": 0.000972565157750343,
|
| 64 |
+
"loss": 0.42,
|
| 65 |
+
"step": 4000
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 6.172839506172839,
|
| 69 |
+
"grad_norm": 0.5927340984344482,
|
| 70 |
+
"learning_rate": 0.0009691358024691358,
|
| 71 |
+
"loss": 0.3612,
|
| 72 |
+
"step": 4500
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 6.858710562414267,
|
| 76 |
+
"grad_norm": 0.5274905562400818,
|
| 77 |
+
"learning_rate": 0.0009657064471879287,
|
| 78 |
+
"loss": 0.2707,
|
| 79 |
+
"step": 5000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 7.544581618655693,
|
| 83 |
+
"grad_norm": 0.46353018283843994,
|
| 84 |
+
"learning_rate": 0.0009622770919067215,
|
| 85 |
+
"loss": 0.2088,
|
| 86 |
+
"step": 5500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 8.23045267489712,
|
| 90 |
+
"grad_norm": 0.7881788611412048,
|
| 91 |
+
"learning_rate": 0.0009588477366255144,
|
| 92 |
+
"loss": 0.2041,
|
| 93 |
+
"step": 6000
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 8.916323731138545,
|
| 97 |
+
"grad_norm": 0.6398069262504578,
|
| 98 |
+
"learning_rate": 0.0009554183813443072,
|
| 99 |
+
"loss": 0.1744,
|
| 100 |
+
"step": 6500
|
| 101 |
+
},
|
| 102 |
+
{
|
| 103 |
+
"epoch": 9.602194787379972,
|
| 104 |
+
"grad_norm": 0.44406837224960327,
|
| 105 |
+
"learning_rate": 0.0009519890260631001,
|
| 106 |
+
"loss": 0.1369,
|
| 107 |
+
"step": 7000
|
| 108 |
+
},
|
| 109 |
+
{
|
| 110 |
+
"epoch": 10.2880658436214,
|
| 111 |
+
"grad_norm": 0.3870859742164612,
|
| 112 |
+
"learning_rate": 0.000948559670781893,
|
| 113 |
+
"loss": 0.1408,
|
| 114 |
+
"step": 7500
|
| 115 |
+
},
|
| 116 |
+
{
|
| 117 |
+
"epoch": 10.973936899862826,
|
| 118 |
+
"grad_norm": 0.6403707265853882,
|
| 119 |
+
"learning_rate": 0.0009451303155006859,
|
| 120 |
+
"loss": 0.1307,
|
| 121 |
+
"step": 8000
|
| 122 |
+
},
|
| 123 |
+
{
|
| 124 |
+
"epoch": 11.659807956104252,
|
| 125 |
+
"grad_norm": 0.406328022480011,
|
| 126 |
+
"learning_rate": 0.0009417009602194788,
|
| 127 |
+
"loss": 0.1105,
|
| 128 |
+
"step": 8500
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 12.345679012345679,
|
| 132 |
+
"grad_norm": 0.4761105179786682,
|
| 133 |
+
"learning_rate": 0.0009382716049382715,
|
| 134 |
+
"loss": 0.1075,
|
| 135 |
+
"step": 9000
|
| 136 |
+
},
|
| 137 |
+
{
|
| 138 |
+
"epoch": 13.031550068587105,
|
| 139 |
+
"grad_norm": 0.39505085349082947,
|
| 140 |
+
"learning_rate": 0.0009348422496570644,
|
| 141 |
+
"loss": 0.1101,
|
| 142 |
+
"step": 9500
|
| 143 |
+
},
|
| 144 |
+
{
|
| 145 |
+
"epoch": 13.717421124828533,
|
| 146 |
+
"grad_norm": 0.3096862733364105,
|
| 147 |
+
"learning_rate": 0.0009314128943758574,
|
| 148 |
+
"loss": 0.0889,
|
| 149 |
+
"step": 10000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 14.40329218106996,
|
| 153 |
+
"grad_norm": 0.681281328201294,
|
| 154 |
+
"learning_rate": 0.0009279835390946503,
|
| 155 |
+
"loss": 0.097,
|
| 156 |
+
"step": 10500
|
| 157 |
+
},
|
| 158 |
+
{
|
| 159 |
+
"epoch": 15.089163237311386,
|
| 160 |
+
"grad_norm": 0.23753629624843597,
|
| 161 |
+
"learning_rate": 0.0009245541838134432,
|
| 162 |
+
"loss": 0.0927,
|
| 163 |
+
"step": 11000
|
| 164 |
+
},
|
| 165 |
+
{
|
| 166 |
+
"epoch": 15.775034293552812,
|
| 167 |
+
"grad_norm": 0.464749813079834,
|
| 168 |
+
"learning_rate": 0.000921124828532236,
|
| 169 |
+
"loss": 0.0793,
|
| 170 |
+
"step": 11500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 16.46090534979424,
|
| 174 |
+
"grad_norm": 0.3283621668815613,
|
| 175 |
+
"learning_rate": 0.0009176954732510289,
|
| 176 |
+
"loss": 0.0764,
|
| 177 |
+
"step": 12000
|
| 178 |
+
},
|
| 179 |
+
{
|
| 180 |
+
"epoch": 17.146776406035666,
|
| 181 |
+
"grad_norm": 0.297809898853302,
|
| 182 |
+
"learning_rate": 0.0009142661179698217,
|
| 183 |
+
"loss": 0.0788,
|
| 184 |
+
"step": 12500
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 17.83264746227709,
|
| 188 |
+
"grad_norm": 0.37593135237693787,
|
| 189 |
+
"learning_rate": 0.0009108367626886146,
|
| 190 |
+
"loss": 0.0746,
|
| 191 |
+
"step": 13000
|
| 192 |
+
},
|
| 193 |
+
{
|
| 194 |
+
"epoch": 18.51851851851852,
|
| 195 |
+
"grad_norm": 0.3363408148288727,
|
| 196 |
+
"learning_rate": 0.0009074074074074074,
|
| 197 |
+
"loss": 0.0764,
|
| 198 |
+
"step": 13500
|
| 199 |
+
},
|
| 200 |
+
{
|
| 201 |
+
"epoch": 19.204389574759944,
|
| 202 |
+
"grad_norm": 0.27103549242019653,
|
| 203 |
+
"learning_rate": 0.0009039780521262003,
|
| 204 |
+
"loss": 0.0751,
|
| 205 |
+
"step": 14000
|
| 206 |
+
},
|
| 207 |
+
{
|
| 208 |
+
"epoch": 19.89026063100137,
|
| 209 |
+
"grad_norm": 0.23534435033798218,
|
| 210 |
+
"learning_rate": 0.0009005486968449932,
|
| 211 |
+
"loss": 0.0643,
|
| 212 |
+
"step": 14500
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 20.5761316872428,
|
| 216 |
+
"grad_norm": 0.3298335671424866,
|
| 217 |
+
"learning_rate": 0.0008971193415637861,
|
| 218 |
+
"loss": 0.0621,
|
| 219 |
+
"step": 15000
|
| 220 |
+
},
|
| 221 |
+
{
|
| 222 |
+
"epoch": 21.262002743484224,
|
| 223 |
+
"grad_norm": 0.28924617171287537,
|
| 224 |
+
"learning_rate": 0.000893689986282579,
|
| 225 |
+
"loss": 0.0661,
|
| 226 |
+
"step": 15500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 21.947873799725652,
|
| 230 |
+
"grad_norm": 0.34902724623680115,
|
| 231 |
+
"learning_rate": 0.0008902606310013717,
|
| 232 |
+
"loss": 0.0693,
|
| 233 |
+
"step": 16000
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 22.633744855967077,
|
| 237 |
+
"grad_norm": 0.24693663418293,
|
| 238 |
+
"learning_rate": 0.0008868312757201646,
|
| 239 |
+
"loss": 0.0591,
|
| 240 |
+
"step": 16500
|
| 241 |
+
},
|
| 242 |
+
{
|
| 243 |
+
"epoch": 23.319615912208505,
|
| 244 |
+
"grad_norm": 0.2739815413951874,
|
| 245 |
+
"learning_rate": 0.0008834019204389575,
|
| 246 |
+
"loss": 0.0576,
|
| 247 |
+
"step": 17000
|
| 248 |
+
},
|
| 249 |
+
{
|
| 250 |
+
"epoch": 24.005486968449933,
|
| 251 |
+
"grad_norm": 0.36366912722587585,
|
| 252 |
+
"learning_rate": 0.0008799725651577504,
|
| 253 |
+
"loss": 0.0555,
|
| 254 |
+
"step": 17500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 24.691358024691358,
|
| 258 |
+
"grad_norm": 0.3027900159358978,
|
| 259 |
+
"learning_rate": 0.0008765432098765433,
|
| 260 |
+
"loss": 0.0507,
|
| 261 |
+
"step": 18000
|
| 262 |
+
},
|
| 263 |
+
{
|
| 264 |
+
"epoch": 25.377229080932786,
|
| 265 |
+
"grad_norm": 0.18507389724254608,
|
| 266 |
+
"learning_rate": 0.0008731138545953361,
|
| 267 |
+
"loss": 0.0525,
|
| 268 |
+
"step": 18500
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 26.06310013717421,
|
| 272 |
+
"grad_norm": 0.23404183983802795,
|
| 273 |
+
"learning_rate": 0.000869684499314129,
|
| 274 |
+
"loss": 0.0579,
|
| 275 |
+
"step": 19000
|
| 276 |
+
},
|
| 277 |
+
{
|
| 278 |
+
"epoch": 26.74897119341564,
|
| 279 |
+
"grad_norm": 0.3586121201515198,
|
| 280 |
+
"learning_rate": 0.0008662551440329218,
|
| 281 |
+
"loss": 0.048,
|
| 282 |
+
"step": 19500
|
| 283 |
+
},
|
| 284 |
+
{
|
| 285 |
+
"epoch": 27.434842249657063,
|
| 286 |
+
"grad_norm": 0.2888980805873871,
|
| 287 |
+
"learning_rate": 0.0008628257887517147,
|
| 288 |
+
"loss": 0.0515,
|
| 289 |
+
"step": 20000
|
| 290 |
+
},
|
| 291 |
+
{
|
| 292 |
+
"epoch": 28.12071330589849,
|
| 293 |
+
"grad_norm": 0.23876026272773743,
|
| 294 |
+
"learning_rate": 0.0008593964334705075,
|
| 295 |
+
"loss": 0.0491,
|
| 296 |
+
"step": 20500
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 28.80658436213992,
|
| 300 |
+
"grad_norm": 0.2646074891090393,
|
| 301 |
+
"learning_rate": 0.0008559670781893004,
|
| 302 |
+
"loss": 0.0466,
|
| 303 |
+
"step": 21000
|
| 304 |
+
},
|
| 305 |
+
{
|
| 306 |
+
"epoch": 29.492455418381343,
|
| 307 |
+
"grad_norm": 0.1991817206144333,
|
| 308 |
+
"learning_rate": 0.0008525377229080933,
|
| 309 |
+
"loss": 0.0451,
|
| 310 |
+
"step": 21500
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 30.17832647462277,
|
| 314 |
+
"grad_norm": 0.218049556016922,
|
| 315 |
+
"learning_rate": 0.0008491083676268862,
|
| 316 |
+
"loss": 0.0445,
|
| 317 |
+
"step": 22000
|
| 318 |
+
},
|
| 319 |
+
{
|
| 320 |
+
"epoch": 30.864197530864196,
|
| 321 |
+
"grad_norm": 0.2530564069747925,
|
| 322 |
+
"learning_rate": 0.0008456790123456791,
|
| 323 |
+
"loss": 0.0427,
|
| 324 |
+
"step": 22500
|
| 325 |
+
},
|
| 326 |
+
{
|
| 327 |
+
"epoch": 31.550068587105624,
|
| 328 |
+
"grad_norm": 0.2068854421377182,
|
| 329 |
+
"learning_rate": 0.0008422496570644718,
|
| 330 |
+
"loss": 0.0421,
|
| 331 |
+
"step": 23000
|
| 332 |
+
},
|
| 333 |
+
{
|
| 334 |
+
"epoch": 32.23593964334705,
|
| 335 |
+
"grad_norm": 0.18602465093135834,
|
| 336 |
+
"learning_rate": 0.0008388203017832647,
|
| 337 |
+
"loss": 0.0437,
|
| 338 |
+
"step": 23500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 32.92181069958848,
|
| 342 |
+
"grad_norm": 0.21352776885032654,
|
| 343 |
+
"learning_rate": 0.0008353909465020576,
|
| 344 |
+
"loss": 0.0462,
|
| 345 |
+
"step": 24000
|
| 346 |
+
},
|
| 347 |
+
{
|
| 348 |
+
"epoch": 33.6076817558299,
|
| 349 |
+
"grad_norm": 0.2586299777030945,
|
| 350 |
+
"learning_rate": 0.0008319615912208505,
|
| 351 |
+
"loss": 0.0417,
|
| 352 |
+
"step": 24500
|
| 353 |
+
},
|
| 354 |
+
{
|
| 355 |
+
"epoch": 34.29355281207133,
|
| 356 |
+
"grad_norm": 0.2829551696777344,
|
| 357 |
+
"learning_rate": 0.0008285322359396434,
|
| 358 |
+
"loss": 0.0381,
|
| 359 |
+
"step": 25000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 34.97942386831276,
|
| 363 |
+
"grad_norm": 0.23624148964881897,
|
| 364 |
+
"learning_rate": 0.0008251028806584362,
|
| 365 |
+
"loss": 0.0361,
|
| 366 |
+
"step": 25500
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 35.66529492455418,
|
| 370 |
+
"grad_norm": 0.21780389547348022,
|
| 371 |
+
"learning_rate": 0.0008216735253772291,
|
| 372 |
+
"loss": 0.0362,
|
| 373 |
+
"step": 26000
|
| 374 |
+
},
|
| 375 |
+
{
|
| 376 |
+
"epoch": 36.351165980795614,
|
| 377 |
+
"grad_norm": 0.3541390299797058,
|
| 378 |
+
"learning_rate": 0.0008182441700960219,
|
| 379 |
+
"loss": 0.0395,
|
| 380 |
+
"step": 26500
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 37.03703703703704,
|
| 384 |
+
"grad_norm": 0.15323896706104279,
|
| 385 |
+
"learning_rate": 0.0008148148148148148,
|
| 386 |
+
"loss": 0.0382,
|
| 387 |
+
"step": 27000
|
| 388 |
+
},
|
| 389 |
+
{
|
| 390 |
+
"epoch": 37.72290809327846,
|
| 391 |
+
"grad_norm": 0.16792021691799164,
|
| 392 |
+
"learning_rate": 0.0008113854595336076,
|
| 393 |
+
"loss": 0.0319,
|
| 394 |
+
"step": 27500
|
| 395 |
+
},
|
| 396 |
+
{
|
| 397 |
+
"epoch": 38.40877914951989,
|
| 398 |
+
"grad_norm": 0.19167844951152802,
|
| 399 |
+
"learning_rate": 0.0008079561042524005,
|
| 400 |
+
"loss": 0.0367,
|
| 401 |
+
"step": 28000
|
| 402 |
+
},
|
| 403 |
+
{
|
| 404 |
+
"epoch": 39.09465020576132,
|
| 405 |
+
"grad_norm": 0.21890634298324585,
|
| 406 |
+
"learning_rate": 0.0008045267489711934,
|
| 407 |
+
"loss": 0.037,
|
| 408 |
+
"step": 28500
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 39.78052126200274,
|
| 412 |
+
"grad_norm": 0.14012588560581207,
|
| 413 |
+
"learning_rate": 0.0008010973936899864,
|
| 414 |
+
"loss": 0.0325,
|
| 415 |
+
"step": 29000
|
| 416 |
+
},
|
| 417 |
+
{
|
| 418 |
+
"epoch": 40.46639231824417,
|
| 419 |
+
"grad_norm": 1.626105546951294,
|
| 420 |
+
"learning_rate": 0.0007976680384087793,
|
| 421 |
+
"loss": 0.0315,
|
| 422 |
+
"step": 29500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 41.1522633744856,
|
| 426 |
+
"grad_norm": 0.1850096881389618,
|
| 427 |
+
"learning_rate": 0.000794238683127572,
|
| 428 |
+
"loss": 0.0379,
|
| 429 |
+
"step": 30000
|
| 430 |
+
},
|
| 431 |
+
{
|
| 432 |
+
"epoch": 41.838134430727024,
|
| 433 |
+
"grad_norm": 0.20383605360984802,
|
| 434 |
+
"learning_rate": 0.0007908093278463649,
|
| 435 |
+
"loss": 0.0315,
|
| 436 |
+
"step": 30500
|
| 437 |
+
},
|
| 438 |
+
{
|
| 439 |
+
"epoch": 42.52400548696845,
|
| 440 |
+
"grad_norm": 0.1643492877483368,
|
| 441 |
+
"learning_rate": 0.0007873799725651578,
|
| 442 |
+
"loss": 0.0276,
|
| 443 |
+
"step": 31000
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 43.20987654320987,
|
| 447 |
+
"grad_norm": 0.15405911207199097,
|
| 448 |
+
"learning_rate": 0.0007839506172839507,
|
| 449 |
+
"loss": 0.0312,
|
| 450 |
+
"step": 31500
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 43.895747599451305,
|
| 454 |
+
"grad_norm": 0.2370378077030182,
|
| 455 |
+
"learning_rate": 0.0007805212620027436,
|
| 456 |
+
"loss": 0.0337,
|
| 457 |
+
"step": 32000
|
| 458 |
+
},
|
| 459 |
+
{
|
| 460 |
+
"epoch": 44.58161865569273,
|
| 461 |
+
"grad_norm": 0.14176137745380402,
|
| 462 |
+
"learning_rate": 0.0007770919067215364,
|
| 463 |
+
"loss": 0.0283,
|
| 464 |
+
"step": 32500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 45.267489711934154,
|
| 468 |
+
"grad_norm": 0.21307243406772614,
|
| 469 |
+
"learning_rate": 0.0007736625514403293,
|
| 470 |
+
"loss": 0.0277,
|
| 471 |
+
"step": 33000
|
| 472 |
+
},
|
| 473 |
+
{
|
| 474 |
+
"epoch": 45.953360768175585,
|
| 475 |
+
"grad_norm": 0.2646368145942688,
|
| 476 |
+
"learning_rate": 0.0007702331961591221,
|
| 477 |
+
"loss": 0.03,
|
| 478 |
+
"step": 33500
|
| 479 |
+
},
|
| 480 |
+
{
|
| 481 |
+
"epoch": 46.63923182441701,
|
| 482 |
+
"grad_norm": 0.22391417622566223,
|
| 483 |
+
"learning_rate": 0.000766803840877915,
|
| 484 |
+
"loss": 0.0298,
|
| 485 |
+
"step": 34000
|
| 486 |
+
},
|
| 487 |
+
{
|
| 488 |
+
"epoch": 47.325102880658434,
|
| 489 |
+
"grad_norm": 0.15177056193351746,
|
| 490 |
+
"learning_rate": 0.0007633744855967078,
|
| 491 |
+
"loss": 0.0272,
|
| 492 |
+
"step": 34500
|
| 493 |
+
},
|
| 494 |
+
{
|
| 495 |
+
"epoch": 48.010973936899866,
|
| 496 |
+
"grad_norm": 0.20461246371269226,
|
| 497 |
+
"learning_rate": 0.0007599451303155007,
|
| 498 |
+
"loss": 0.0278,
|
| 499 |
+
"step": 35000
|
| 500 |
+
},
|
| 501 |
+
{
|
| 502 |
+
"epoch": 48.69684499314129,
|
| 503 |
+
"grad_norm": 0.12500979006290436,
|
| 504 |
+
"learning_rate": 0.0007565157750342936,
|
| 505 |
+
"loss": 0.0268,
|
| 506 |
+
"step": 35500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 49.382716049382715,
|
| 510 |
+
"grad_norm": 0.4091637134552002,
|
| 511 |
+
"learning_rate": 0.0007530864197530865,
|
| 512 |
+
"loss": 0.0267,
|
| 513 |
+
"step": 36000
|
| 514 |
+
},
|
| 515 |
+
{
|
| 516 |
+
"epoch": 50.06858710562414,
|
| 517 |
+
"grad_norm": 0.22375427186489105,
|
| 518 |
+
"learning_rate": 0.0007496570644718793,
|
| 519 |
+
"loss": 0.0327,
|
| 520 |
+
"step": 36500
|
| 521 |
+
},
|
| 522 |
+
{
|
| 523 |
+
"epoch": 50.75445816186557,
|
| 524 |
+
"grad_norm": 0.11407098174095154,
|
| 525 |
+
"learning_rate": 0.0007462277091906721,
|
| 526 |
+
"loss": 0.0245,
|
| 527 |
+
"step": 37000
|
| 528 |
+
},
|
| 529 |
+
{
|
| 530 |
+
"epoch": 51.440329218106996,
|
| 531 |
+
"grad_norm": 0.1025354415178299,
|
| 532 |
+
"learning_rate": 0.000742798353909465,
|
| 533 |
+
"loss": 0.0246,
|
| 534 |
+
"step": 37500
|
| 535 |
+
},
|
| 536 |
+
{
|
| 537 |
+
"epoch": 52.12620027434842,
|
| 538 |
+
"grad_norm": 0.20578157901763916,
|
| 539 |
+
"learning_rate": 0.0007393689986282579,
|
| 540 |
+
"loss": 0.0258,
|
| 541 |
+
"step": 38000
|
| 542 |
+
},
|
| 543 |
+
{
|
| 544 |
+
"epoch": 52.81207133058985,
|
| 545 |
+
"grad_norm": 0.1874309480190277,
|
| 546 |
+
"learning_rate": 0.0007359396433470508,
|
| 547 |
+
"loss": 0.0276,
|
| 548 |
+
"step": 38500
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 53.49794238683128,
|
| 552 |
+
"grad_norm": 0.1793011724948883,
|
| 553 |
+
"learning_rate": 0.0007325102880658437,
|
| 554 |
+
"loss": 0.0232,
|
| 555 |
+
"step": 39000
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 54.1838134430727,
|
| 559 |
+
"grad_norm": 0.18910464644432068,
|
| 560 |
+
"learning_rate": 0.0007290809327846365,
|
| 561 |
+
"loss": 0.0237,
|
| 562 |
+
"step": 39500
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 54.869684499314126,
|
| 566 |
+
"grad_norm": 0.26874855160713196,
|
| 567 |
+
"learning_rate": 0.0007256515775034293,
|
| 568 |
+
"loss": 0.0249,
|
| 569 |
+
"step": 40000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 55.55555555555556,
|
| 573 |
+
"grad_norm": 0.12961339950561523,
|
| 574 |
+
"learning_rate": 0.0007222222222222222,
|
| 575 |
+
"loss": 0.0262,
|
| 576 |
+
"step": 40500
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 56.24142661179698,
|
| 580 |
+
"grad_norm": 0.14723250269889832,
|
| 581 |
+
"learning_rate": 0.0007187928669410151,
|
| 582 |
+
"loss": 0.0224,
|
| 583 |
+
"step": 41000
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 56.927297668038406,
|
| 587 |
+
"grad_norm": 0.17689082026481628,
|
| 588 |
+
"learning_rate": 0.000715363511659808,
|
| 589 |
+
"loss": 0.0226,
|
| 590 |
+
"step": 41500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 57.61316872427984,
|
| 594 |
+
"grad_norm": 0.14181144535541534,
|
| 595 |
+
"learning_rate": 0.0007119341563786008,
|
| 596 |
+
"loss": 0.0214,
|
| 597 |
+
"step": 42000
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 58.29903978052126,
|
| 601 |
+
"grad_norm": 0.5437673330307007,
|
| 602 |
+
"learning_rate": 0.0007085048010973937,
|
| 603 |
+
"loss": 0.0234,
|
| 604 |
+
"step": 42500
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 58.98491083676269,
|
| 608 |
+
"grad_norm": 0.0570339597761631,
|
| 609 |
+
"learning_rate": 0.0007050754458161866,
|
| 610 |
+
"loss": 0.0238,
|
| 611 |
+
"step": 43000
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 59.67078189300412,
|
| 615 |
+
"grad_norm": 0.14648722112178802,
|
| 616 |
+
"learning_rate": 0.0007016460905349794,
|
| 617 |
+
"loss": 0.0209,
|
| 618 |
+
"step": 43500
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 60.35665294924554,
|
| 622 |
+
"grad_norm": 0.06610771268606186,
|
| 623 |
+
"learning_rate": 0.0006982167352537722,
|
| 624 |
+
"loss": 0.0203,
|
| 625 |
+
"step": 44000
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 61.04252400548697,
|
| 629 |
+
"grad_norm": 0.092800073325634,
|
| 630 |
+
"learning_rate": 0.0006947873799725651,
|
| 631 |
+
"loss": 0.0198,
|
| 632 |
+
"step": 44500
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 61.72839506172839,
|
| 636 |
+
"grad_norm": 0.12414117157459259,
|
| 637 |
+
"learning_rate": 0.000691358024691358,
|
| 638 |
+
"loss": 0.0195,
|
| 639 |
+
"step": 45000
|
| 640 |
+
},
|
| 641 |
+
{
|
| 642 |
+
"epoch": 62.414266117969824,
|
| 643 |
+
"grad_norm": 0.10231557488441467,
|
| 644 |
+
"learning_rate": 0.0006879286694101509,
|
| 645 |
+
"loss": 0.0225,
|
| 646 |
+
"step": 45500
|
| 647 |
+
},
|
| 648 |
+
{
|
| 649 |
+
"epoch": 63.10013717421125,
|
| 650 |
+
"grad_norm": 0.14670057594776154,
|
| 651 |
+
"learning_rate": 0.0006844993141289438,
|
| 652 |
+
"loss": 0.0213,
|
| 653 |
+
"step": 46000
|
| 654 |
+
},
|
| 655 |
+
{
|
| 656 |
+
"epoch": 63.78600823045267,
|
| 657 |
+
"grad_norm": 0.1486993432044983,
|
| 658 |
+
"learning_rate": 0.0006810699588477366,
|
| 659 |
+
"loss": 0.0225,
|
| 660 |
+
"step": 46500
|
| 661 |
+
},
|
| 662 |
+
{
|
| 663 |
+
"epoch": 64.4718792866941,
|
| 664 |
+
"grad_norm": 0.11352519690990448,
|
| 665 |
+
"learning_rate": 0.0006776406035665294,
|
| 666 |
+
"loss": 0.0187,
|
| 667 |
+
"step": 47000
|
| 668 |
+
},
|
| 669 |
+
{
|
| 670 |
+
"epoch": 65.15775034293553,
|
| 671 |
+
"grad_norm": 0.08721095323562622,
|
| 672 |
+
"learning_rate": 0.0006742112482853223,
|
| 673 |
+
"loss": 0.0186,
|
| 674 |
+
"step": 47500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 65.84362139917695,
|
| 678 |
+
"grad_norm": 0.2794257402420044,
|
| 679 |
+
"learning_rate": 0.0006707818930041153,
|
| 680 |
+
"loss": 0.0184,
|
| 681 |
+
"step": 48000
|
| 682 |
+
},
|
| 683 |
+
{
|
| 684 |
+
"epoch": 66.52949245541838,
|
| 685 |
+
"grad_norm": 0.16597363352775574,
|
| 686 |
+
"learning_rate": 0.0006673525377229082,
|
| 687 |
+
"loss": 0.0203,
|
| 688 |
+
"step": 48500
|
| 689 |
+
},
|
| 690 |
+
{
|
| 691 |
+
"epoch": 67.2153635116598,
|
| 692 |
+
"grad_norm": 0.1067240834236145,
|
| 693 |
+
"learning_rate": 0.000663923182441701,
|
| 694 |
+
"loss": 0.0208,
|
| 695 |
+
"step": 49000
|
| 696 |
+
},
|
| 697 |
+
{
|
| 698 |
+
"epoch": 67.90123456790124,
|
| 699 |
+
"grad_norm": 0.13352781534194946,
|
| 700 |
+
"learning_rate": 0.0006604938271604939,
|
| 701 |
+
"loss": 0.0181,
|
| 702 |
+
"step": 49500
|
| 703 |
+
},
|
| 704 |
+
{
|
| 705 |
+
"epoch": 68.58710562414267,
|
| 706 |
+
"grad_norm": 0.15161901712417603,
|
| 707 |
+
"learning_rate": 0.0006570644718792868,
|
| 708 |
+
"loss": 0.02,
|
| 709 |
+
"step": 50000
|
| 710 |
+
},
|
| 711 |
+
{
|
| 712 |
+
"epoch": 69.27297668038409,
|
| 713 |
+
"grad_norm": 0.19234149158000946,
|
| 714 |
+
"learning_rate": 0.0006536351165980796,
|
| 715 |
+
"loss": 0.0179,
|
| 716 |
+
"step": 50500
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 69.95884773662551,
|
| 720 |
+
"grad_norm": 0.11277095228433609,
|
| 721 |
+
"learning_rate": 0.0006502057613168724,
|
| 722 |
+
"loss": 0.0181,
|
| 723 |
+
"step": 51000
|
| 724 |
+
},
|
| 725 |
+
{
|
| 726 |
+
"epoch": 70.64471879286694,
|
| 727 |
+
"grad_norm": 0.10098372399806976,
|
| 728 |
+
"learning_rate": 0.0006467764060356653,
|
| 729 |
+
"loss": 0.0187,
|
| 730 |
+
"step": 51500
|
| 731 |
+
},
|
| 732 |
+
{
|
| 733 |
+
"epoch": 71.33058984910836,
|
| 734 |
+
"grad_norm": 0.08452719449996948,
|
| 735 |
+
"learning_rate": 0.0006433470507544582,
|
| 736 |
+
"loss": 0.0167,
|
| 737 |
+
"step": 52000
|
| 738 |
+
},
|
| 739 |
+
{
|
| 740 |
+
"epoch": 72.01646090534979,
|
| 741 |
+
"grad_norm": 0.11851578205823898,
|
| 742 |
+
"learning_rate": 0.0006399176954732511,
|
| 743 |
+
"loss": 0.0189,
|
| 744 |
+
"step": 52500
|
| 745 |
+
},
|
| 746 |
+
{
|
| 747 |
+
"epoch": 72.70233196159123,
|
| 748 |
+
"grad_norm": 0.2409876585006714,
|
| 749 |
+
"learning_rate": 0.000636488340192044,
|
| 750 |
+
"loss": 0.018,
|
| 751 |
+
"step": 53000
|
| 752 |
+
},
|
| 753 |
+
{
|
| 754 |
+
"epoch": 73.38820301783265,
|
| 755 |
+
"grad_norm": 0.1037708967924118,
|
| 756 |
+
"learning_rate": 0.0006330589849108368,
|
| 757 |
+
"loss": 0.0159,
|
| 758 |
+
"step": 53500
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 74.07407407407408,
|
| 762 |
+
"grad_norm": 0.1006944552063942,
|
| 763 |
+
"learning_rate": 0.0006296296296296296,
|
| 764 |
+
"loss": 0.018,
|
| 765 |
+
"step": 54000
|
| 766 |
+
},
|
| 767 |
+
{
|
| 768 |
+
"epoch": 74.7599451303155,
|
| 769 |
+
"grad_norm": 0.11186862736940384,
|
| 770 |
+
"learning_rate": 0.0006262002743484225,
|
| 771 |
+
"loss": 0.0169,
|
| 772 |
+
"step": 54500
|
| 773 |
+
},
|
| 774 |
+
{
|
| 775 |
+
"epoch": 75.44581618655693,
|
| 776 |
+
"grad_norm": 0.08759860694408417,
|
| 777 |
+
"learning_rate": 0.0006227709190672154,
|
| 778 |
+
"loss": 0.0165,
|
| 779 |
+
"step": 55000
|
| 780 |
+
},
|
| 781 |
+
{
|
| 782 |
+
"epoch": 76.13168724279835,
|
| 783 |
+
"grad_norm": 0.14316201210021973,
|
| 784 |
+
"learning_rate": 0.0006193415637860083,
|
| 785 |
+
"loss": 0.0176,
|
| 786 |
+
"step": 55500
|
| 787 |
+
},
|
| 788 |
+
{
|
| 789 |
+
"epoch": 76.81755829903977,
|
| 790 |
+
"grad_norm": 0.1392996907234192,
|
| 791 |
+
"learning_rate": 0.0006159122085048011,
|
| 792 |
+
"loss": 0.0161,
|
| 793 |
+
"step": 56000
|
| 794 |
+
},
|
| 795 |
+
{
|
| 796 |
+
"epoch": 77.50342935528121,
|
| 797 |
+
"grad_norm": 0.14493365585803986,
|
| 798 |
+
"learning_rate": 0.000612482853223594,
|
| 799 |
+
"loss": 0.0157,
|
| 800 |
+
"step": 56500
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 78.18930041152264,
|
| 804 |
+
"grad_norm": 0.0993877425789833,
|
| 805 |
+
"learning_rate": 0.0006090534979423869,
|
| 806 |
+
"loss": 0.0168,
|
| 807 |
+
"step": 57000
|
| 808 |
+
},
|
| 809 |
+
{
|
| 810 |
+
"epoch": 78.87517146776406,
|
| 811 |
+
"grad_norm": 0.09831307828426361,
|
| 812 |
+
"learning_rate": 0.0006056241426611797,
|
| 813 |
+
"loss": 0.0164,
|
| 814 |
+
"step": 57500
|
| 815 |
+
},
|
| 816 |
+
{
|
| 817 |
+
"epoch": 79.56104252400549,
|
| 818 |
+
"grad_norm": 0.09205558151006699,
|
| 819 |
+
"learning_rate": 0.0006021947873799725,
|
| 820 |
+
"loss": 0.0157,
|
| 821 |
+
"step": 58000
|
| 822 |
+
},
|
| 823 |
+
{
|
| 824 |
+
"epoch": 80.24691358024691,
|
| 825 |
+
"grad_norm": 0.10848256945610046,
|
| 826 |
+
"learning_rate": 0.0005987654320987654,
|
| 827 |
+
"loss": 0.014,
|
| 828 |
+
"step": 58500
|
| 829 |
+
},
|
| 830 |
+
{
|
| 831 |
+
"epoch": 80.93278463648834,
|
| 832 |
+
"grad_norm": 0.14594706892967224,
|
| 833 |
+
"learning_rate": 0.0005953360768175583,
|
| 834 |
+
"loss": 0.0144,
|
| 835 |
+
"step": 59000
|
| 836 |
+
},
|
| 837 |
+
{
|
| 838 |
+
"epoch": 81.61865569272976,
|
| 839 |
+
"grad_norm": 0.057823847979307175,
|
| 840 |
+
"learning_rate": 0.0005919067215363512,
|
| 841 |
+
"loss": 0.0143,
|
| 842 |
+
"step": 59500
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 82.3045267489712,
|
| 846 |
+
"grad_norm": 0.3916853368282318,
|
| 847 |
+
"learning_rate": 0.0005884773662551441,
|
| 848 |
+
"loss": 0.0159,
|
| 849 |
+
"step": 60000
|
| 850 |
+
},
|
| 851 |
+
{
|
| 852 |
+
"epoch": 82.99039780521262,
|
| 853 |
+
"grad_norm": 0.11421903222799301,
|
| 854 |
+
"learning_rate": 0.0005850480109739369,
|
| 855 |
+
"loss": 0.017,
|
| 856 |
+
"step": 60500
|
| 857 |
+
},
|
| 858 |
+
{
|
| 859 |
+
"epoch": 83.67626886145405,
|
| 860 |
+
"grad_norm": 0.10148236900568008,
|
| 861 |
+
"learning_rate": 0.0005816186556927297,
|
| 862 |
+
"loss": 0.014,
|
| 863 |
+
"step": 61000
|
| 864 |
+
},
|
| 865 |
+
{
|
| 866 |
+
"epoch": 84.36213991769547,
|
| 867 |
+
"grad_norm": 0.12611427903175354,
|
| 868 |
+
"learning_rate": 0.0005781893004115226,
|
| 869 |
+
"loss": 0.0139,
|
| 870 |
+
"step": 61500
|
| 871 |
+
},
|
| 872 |
+
{
|
| 873 |
+
"epoch": 85.0480109739369,
|
| 874 |
+
"grad_norm": 0.07960600405931473,
|
| 875 |
+
"learning_rate": 0.0005747599451303155,
|
| 876 |
+
"loss": 0.0145,
|
| 877 |
+
"step": 62000
|
| 878 |
+
},
|
| 879 |
+
{
|
| 880 |
+
"epoch": 85.73388203017832,
|
| 881 |
+
"grad_norm": 0.09838591516017914,
|
| 882 |
+
"learning_rate": 0.0005713305898491084,
|
| 883 |
+
"loss": 0.0131,
|
| 884 |
+
"step": 62500
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 86.41975308641975,
|
| 888 |
+
"grad_norm": 0.06399769335985184,
|
| 889 |
+
"learning_rate": 0.0005679012345679012,
|
| 890 |
+
"loss": 0.0131,
|
| 891 |
+
"step": 63000
|
| 892 |
+
},
|
| 893 |
+
{
|
| 894 |
+
"epoch": 87.10562414266118,
|
| 895 |
+
"grad_norm": 0.1615062952041626,
|
| 896 |
+
"learning_rate": 0.0005644718792866941,
|
| 897 |
+
"loss": 0.0159,
|
| 898 |
+
"step": 63500
|
| 899 |
+
},
|
| 900 |
+
{
|
| 901 |
+
"epoch": 87.79149519890261,
|
| 902 |
+
"grad_norm": 0.11928918957710266,
|
| 903 |
+
"learning_rate": 0.000561042524005487,
|
| 904 |
+
"loss": 0.0153,
|
| 905 |
+
"step": 64000
|
| 906 |
+
},
|
| 907 |
+
{
|
| 908 |
+
"epoch": 88.47736625514403,
|
| 909 |
+
"grad_norm": 0.0789957344532013,
|
| 910 |
+
"learning_rate": 0.0005576131687242798,
|
| 911 |
+
"loss": 0.0128,
|
| 912 |
+
"step": 64500
|
| 913 |
+
},
|
| 914 |
+
{
|
| 915 |
+
"epoch": 89.16323731138546,
|
| 916 |
+
"grad_norm": 0.10119163990020752,
|
| 917 |
+
"learning_rate": 0.0005541838134430726,
|
| 918 |
+
"loss": 0.0134,
|
| 919 |
+
"step": 65000
|
| 920 |
+
},
|
| 921 |
+
{
|
| 922 |
+
"epoch": 89.84910836762688,
|
| 923 |
+
"grad_norm": 0.10709987580776215,
|
| 924 |
+
"learning_rate": 0.0005507544581618655,
|
| 925 |
+
"loss": 0.0129,
|
| 926 |
+
"step": 65500
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 90.53497942386831,
|
| 930 |
+
"grad_norm": 0.1411323994398117,
|
| 931 |
+
"learning_rate": 0.0005473251028806584,
|
| 932 |
+
"loss": 0.0135,
|
| 933 |
+
"step": 66000
|
| 934 |
+
},
|
| 935 |
+
{
|
| 936 |
+
"epoch": 91.22085048010975,
|
| 937 |
+
"grad_norm": 0.1025325134396553,
|
| 938 |
+
"learning_rate": 0.0005438957475994513,
|
| 939 |
+
"loss": 0.0128,
|
| 940 |
+
"step": 66500
|
| 941 |
+
},
|
| 942 |
+
{
|
| 943 |
+
"epoch": 91.90672153635117,
|
| 944 |
+
"grad_norm": 0.11264779418706894,
|
| 945 |
+
"learning_rate": 0.0005404663923182442,
|
| 946 |
+
"loss": 0.0143,
|
| 947 |
+
"step": 67000
|
| 948 |
+
},
|
| 949 |
+
{
|
| 950 |
+
"epoch": 92.5925925925926,
|
| 951 |
+
"grad_norm": 0.11864063143730164,
|
| 952 |
+
"learning_rate": 0.0005370370370370371,
|
| 953 |
+
"loss": 0.0124,
|
| 954 |
+
"step": 67500
|
| 955 |
+
},
|
| 956 |
+
{
|
| 957 |
+
"epoch": 93.27846364883402,
|
| 958 |
+
"grad_norm": 0.07646331936120987,
|
| 959 |
+
"learning_rate": 0.0005336076817558299,
|
| 960 |
+
"loss": 0.0126,
|
| 961 |
+
"step": 68000
|
| 962 |
+
},
|
| 963 |
+
{
|
| 964 |
+
"epoch": 93.96433470507544,
|
| 965 |
+
"grad_norm": 0.06260576099157333,
|
| 966 |
+
"learning_rate": 0.0005301783264746228,
|
| 967 |
+
"loss": 0.0116,
|
| 968 |
+
"step": 68500
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 94.65020576131687,
|
| 972 |
+
"grad_norm": 0.10267277806997299,
|
| 973 |
+
"learning_rate": 0.0005267489711934157,
|
| 974 |
+
"loss": 0.0117,
|
| 975 |
+
"step": 69000
|
| 976 |
+
},
|
| 977 |
+
{
|
| 978 |
+
"epoch": 95.3360768175583,
|
| 979 |
+
"grad_norm": 0.07879356294870377,
|
| 980 |
+
"learning_rate": 0.0005233196159122086,
|
| 981 |
+
"loss": 0.0133,
|
| 982 |
+
"step": 69500
|
| 983 |
+
},
|
| 984 |
+
{
|
| 985 |
+
"epoch": 96.02194787379973,
|
| 986 |
+
"grad_norm": 0.06801025569438934,
|
| 987 |
+
"learning_rate": 0.0005198902606310014,
|
| 988 |
+
"loss": 0.0121,
|
| 989 |
+
"step": 70000
|
| 990 |
+
},
|
| 991 |
+
{
|
| 992 |
+
"epoch": 96.70781893004116,
|
| 993 |
+
"grad_norm": 0.1383305788040161,
|
| 994 |
+
"learning_rate": 0.0005164609053497943,
|
| 995 |
+
"loss": 0.0118,
|
| 996 |
+
"step": 70500
|
| 997 |
+
},
|
| 998 |
+
{
|
| 999 |
+
"epoch": 97.39368998628258,
|
| 1000 |
+
"grad_norm": 0.13461919128894806,
|
| 1001 |
+
"learning_rate": 0.0005130315500685872,
|
| 1002 |
+
"loss": 0.0121,
|
| 1003 |
+
"step": 71000
|
| 1004 |
+
},
|
| 1005 |
+
{
|
| 1006 |
+
"epoch": 98.079561042524,
|
| 1007 |
+
"grad_norm": 0.07046571373939514,
|
| 1008 |
+
"learning_rate": 0.00050960219478738,
|
| 1009 |
+
"loss": 0.0119,
|
| 1010 |
+
"step": 71500
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 98.76543209876543,
|
| 1014 |
+
"grad_norm": 0.12527473270893097,
|
| 1015 |
+
"learning_rate": 0.0005061728395061728,
|
| 1016 |
+
"loss": 0.0116,
|
| 1017 |
+
"step": 72000
|
| 1018 |
+
},
|
| 1019 |
+
{
|
| 1020 |
+
"epoch": 99.45130315500685,
|
| 1021 |
+
"grad_norm": 0.08155812323093414,
|
| 1022 |
+
"learning_rate": 0.0005027434842249657,
|
| 1023 |
+
"loss": 0.0108,
|
| 1024 |
+
"step": 72500
|
| 1025 |
+
},
|
| 1026 |
+
{
|
| 1027 |
+
"epoch": 100.13717421124828,
|
| 1028 |
+
"grad_norm": 0.10491594672203064,
|
| 1029 |
+
"learning_rate": 0.0004993141289437586,
|
| 1030 |
+
"loss": 0.0112,
|
| 1031 |
+
"step": 73000
|
| 1032 |
+
},
|
| 1033 |
+
{
|
| 1034 |
+
"epoch": 100.82304526748972,
|
| 1035 |
+
"grad_norm": 0.10411892831325531,
|
| 1036 |
+
"learning_rate": 0.0004958847736625515,
|
| 1037 |
+
"loss": 0.0105,
|
| 1038 |
+
"step": 73500
|
| 1039 |
+
},
|
| 1040 |
+
{
|
| 1041 |
+
"epoch": 101.50891632373114,
|
| 1042 |
+
"grad_norm": 0.049623072147369385,
|
| 1043 |
+
"learning_rate": 0.0004924554183813444,
|
| 1044 |
+
"loss": 0.0111,
|
| 1045 |
+
"step": 74000
|
| 1046 |
+
},
|
| 1047 |
+
{
|
| 1048 |
+
"epoch": 102.19478737997257,
|
| 1049 |
+
"grad_norm": 0.11287267506122589,
|
| 1050 |
+
"learning_rate": 0.0004890260631001372,
|
| 1051 |
+
"loss": 0.0113,
|
| 1052 |
+
"step": 74500
|
| 1053 |
+
},
|
| 1054 |
+
{
|
| 1055 |
+
"epoch": 102.88065843621399,
|
| 1056 |
+
"grad_norm": 0.09406940639019012,
|
| 1057 |
+
"learning_rate": 0.00048559670781893007,
|
| 1058 |
+
"loss": 0.0113,
|
| 1059 |
+
"step": 75000
|
| 1060 |
+
},
|
| 1061 |
+
{
|
| 1062 |
+
"epoch": 103.56652949245542,
|
| 1063 |
+
"grad_norm": 0.05741545557975769,
|
| 1064 |
+
"learning_rate": 0.00048216735253772295,
|
| 1065 |
+
"loss": 0.0116,
|
| 1066 |
+
"step": 75500
|
| 1067 |
+
},
|
| 1068 |
+
{
|
| 1069 |
+
"epoch": 104.25240054869684,
|
| 1070 |
+
"grad_norm": 0.06398554146289825,
|
| 1071 |
+
"learning_rate": 0.0004787379972565158,
|
| 1072 |
+
"loss": 0.0124,
|
| 1073 |
+
"step": 76000
|
| 1074 |
+
},
|
| 1075 |
+
{
|
| 1076 |
+
"epoch": 104.93827160493827,
|
| 1077 |
+
"grad_norm": 0.05331519991159439,
|
| 1078 |
+
"learning_rate": 0.00047530864197530866,
|
| 1079 |
+
"loss": 0.0105,
|
| 1080 |
+
"step": 76500
|
| 1081 |
+
},
|
| 1082 |
+
{
|
| 1083 |
+
"epoch": 105.6241426611797,
|
| 1084 |
+
"grad_norm": 0.040314387530088425,
|
| 1085 |
+
"learning_rate": 0.00047187928669410154,
|
| 1086 |
+
"loss": 0.0091,
|
| 1087 |
+
"step": 77000
|
| 1088 |
+
},
|
| 1089 |
+
{
|
| 1090 |
+
"epoch": 106.31001371742113,
|
| 1091 |
+
"grad_norm": 0.0661318302154541,
|
| 1092 |
+
"learning_rate": 0.00046844993141289436,
|
| 1093 |
+
"loss": 0.0096,
|
| 1094 |
+
"step": 77500
|
| 1095 |
+
},
|
| 1096 |
+
{
|
| 1097 |
+
"epoch": 106.99588477366255,
|
| 1098 |
+
"grad_norm": 0.13146652281284332,
|
| 1099 |
+
"learning_rate": 0.00046502057613168724,
|
| 1100 |
+
"loss": 0.0112,
|
| 1101 |
+
"step": 78000
|
| 1102 |
+
},
|
| 1103 |
+
{
|
| 1104 |
+
"epoch": 107.68175582990398,
|
| 1105 |
+
"grad_norm": 0.060037847608327866,
|
| 1106 |
+
"learning_rate": 0.0004615912208504801,
|
| 1107 |
+
"loss": 0.0096,
|
| 1108 |
+
"step": 78500
|
| 1109 |
+
},
|
| 1110 |
+
{
|
| 1111 |
+
"epoch": 108.3676268861454,
|
| 1112 |
+
"grad_norm": 0.10393790900707245,
|
| 1113 |
+
"learning_rate": 0.00045816186556927295,
|
| 1114 |
+
"loss": 0.0107,
|
| 1115 |
+
"step": 79000
|
| 1116 |
+
},
|
| 1117 |
+
{
|
| 1118 |
+
"epoch": 109.05349794238683,
|
| 1119 |
+
"grad_norm": 0.053112734109163284,
|
| 1120 |
+
"learning_rate": 0.0004547325102880658,
|
| 1121 |
+
"loss": 0.0096,
|
| 1122 |
+
"step": 79500
|
| 1123 |
+
},
|
| 1124 |
+
{
|
| 1125 |
+
"epoch": 109.73936899862825,
|
| 1126 |
+
"grad_norm": 0.05577271804213524,
|
| 1127 |
+
"learning_rate": 0.0004513031550068587,
|
| 1128 |
+
"loss": 0.0091,
|
| 1129 |
+
"step": 80000
|
| 1130 |
+
},
|
| 1131 |
+
{
|
| 1132 |
+
"epoch": 110.42524005486969,
|
| 1133 |
+
"grad_norm": 0.03647785261273384,
|
| 1134 |
+
"learning_rate": 0.0004478737997256516,
|
| 1135 |
+
"loss": 0.009,
|
| 1136 |
+
"step": 80500
|
| 1137 |
+
},
|
| 1138 |
+
{
|
| 1139 |
+
"epoch": 111.11111111111111,
|
| 1140 |
+
"grad_norm": 0.09830909222364426,
|
| 1141 |
+
"learning_rate": 0.0004444444444444444,
|
| 1142 |
+
"loss": 0.0098,
|
| 1143 |
+
"step": 81000
|
| 1144 |
+
},
|
| 1145 |
+
{
|
| 1146 |
+
"epoch": 111.79698216735254,
|
| 1147 |
+
"grad_norm": 0.025291450321674347,
|
| 1148 |
+
"learning_rate": 0.0004410150891632373,
|
| 1149 |
+
"loss": 0.0099,
|
| 1150 |
+
"step": 81500
|
| 1151 |
+
},
|
| 1152 |
+
{
|
| 1153 |
+
"epoch": 112.48285322359396,
|
| 1154 |
+
"grad_norm": 0.0518038235604763,
|
| 1155 |
+
"learning_rate": 0.0004375857338820302,
|
| 1156 |
+
"loss": 0.0093,
|
| 1157 |
+
"step": 82000
|
| 1158 |
+
},
|
| 1159 |
+
{
|
| 1160 |
+
"epoch": 113.16872427983539,
|
| 1161 |
+
"grad_norm": 0.08746583759784698,
|
| 1162 |
+
"learning_rate": 0.00043415637860082305,
|
| 1163 |
+
"loss": 0.0096,
|
| 1164 |
+
"step": 82500
|
| 1165 |
+
},
|
| 1166 |
+
{
|
| 1167 |
+
"epoch": 113.85459533607681,
|
| 1168 |
+
"grad_norm": 0.0944758877158165,
|
| 1169 |
+
"learning_rate": 0.00043072702331961593,
|
| 1170 |
+
"loss": 0.0098,
|
| 1171 |
+
"step": 83000
|
| 1172 |
+
},
|
| 1173 |
+
{
|
| 1174 |
+
"epoch": 114.54046639231825,
|
| 1175 |
+
"grad_norm": 0.10082614421844482,
|
| 1176 |
+
"learning_rate": 0.0004272976680384088,
|
| 1177 |
+
"loss": 0.0087,
|
| 1178 |
+
"step": 83500
|
| 1179 |
+
},
|
| 1180 |
+
{
|
| 1181 |
+
"epoch": 115.22633744855968,
|
| 1182 |
+
"grad_norm": 0.05366198346018791,
|
| 1183 |
+
"learning_rate": 0.0004238683127572017,
|
| 1184 |
+
"loss": 0.0088,
|
| 1185 |
+
"step": 84000
|
| 1186 |
+
},
|
| 1187 |
+
{
|
| 1188 |
+
"epoch": 115.9122085048011,
|
| 1189 |
+
"grad_norm": 0.04216461256146431,
|
| 1190 |
+
"learning_rate": 0.0004204389574759945,
|
| 1191 |
+
"loss": 0.0083,
|
| 1192 |
+
"step": 84500
|
| 1193 |
+
},
|
| 1194 |
+
{
|
| 1195 |
+
"epoch": 116.59807956104252,
|
| 1196 |
+
"grad_norm": 0.061591554433107376,
|
| 1197 |
+
"learning_rate": 0.0004170096021947874,
|
| 1198 |
+
"loss": 0.0098,
|
| 1199 |
+
"step": 85000
|
| 1200 |
+
},
|
| 1201 |
+
{
|
| 1202 |
+
"epoch": 117.28395061728395,
|
| 1203 |
+
"grad_norm": 0.10498243570327759,
|
| 1204 |
+
"learning_rate": 0.0004135802469135803,
|
| 1205 |
+
"loss": 0.0093,
|
| 1206 |
+
"step": 85500
|
| 1207 |
+
},
|
| 1208 |
+
{
|
| 1209 |
+
"epoch": 117.96982167352537,
|
| 1210 |
+
"grad_norm": 0.041311949491500854,
|
| 1211 |
+
"learning_rate": 0.0004101508916323731,
|
| 1212 |
+
"loss": 0.0087,
|
| 1213 |
+
"step": 86000
|
| 1214 |
+
},
|
| 1215 |
+
{
|
| 1216 |
+
"epoch": 118.6556927297668,
|
| 1217 |
+
"grad_norm": 0.09587587416172028,
|
| 1218 |
+
"learning_rate": 0.000406721536351166,
|
| 1219 |
+
"loss": 0.0098,
|
| 1220 |
+
"step": 86500
|
| 1221 |
+
},
|
| 1222 |
+
{
|
| 1223 |
+
"epoch": 119.34156378600824,
|
| 1224 |
+
"grad_norm": 0.05686916410923004,
|
| 1225 |
+
"learning_rate": 0.00040329218106995886,
|
| 1226 |
+
"loss": 0.0084,
|
| 1227 |
+
"step": 87000
|
| 1228 |
+
},
|
| 1229 |
+
{
|
| 1230 |
+
"epoch": 120.02743484224966,
|
| 1231 |
+
"grad_norm": 0.06297193467617035,
|
| 1232 |
+
"learning_rate": 0.00039986282578875174,
|
| 1233 |
+
"loss": 0.0082,
|
| 1234 |
+
"step": 87500
|
| 1235 |
+
},
|
| 1236 |
+
{
|
| 1237 |
+
"epoch": 120.71330589849109,
|
| 1238 |
+
"grad_norm": 0.13432051241397858,
|
| 1239 |
+
"learning_rate": 0.00039643347050754456,
|
| 1240 |
+
"loss": 0.0075,
|
| 1241 |
+
"step": 88000
|
| 1242 |
+
},
|
| 1243 |
+
{
|
| 1244 |
+
"epoch": 121.39917695473251,
|
| 1245 |
+
"grad_norm": 0.03966566175222397,
|
| 1246 |
+
"learning_rate": 0.00039300411522633744,
|
| 1247 |
+
"loss": 0.0133,
|
| 1248 |
+
"step": 88500
|
| 1249 |
+
},
|
| 1250 |
+
{
|
| 1251 |
+
"epoch": 122.08504801097394,
|
| 1252 |
+
"grad_norm": 0.09373347461223602,
|
| 1253 |
+
"learning_rate": 0.0003895747599451303,
|
| 1254 |
+
"loss": 0.0082,
|
| 1255 |
+
"step": 89000
|
| 1256 |
+
},
|
| 1257 |
+
{
|
| 1258 |
+
"epoch": 122.77091906721536,
|
| 1259 |
+
"grad_norm": 0.06179581582546234,
|
| 1260 |
+
"learning_rate": 0.00038614540466392315,
|
| 1261 |
+
"loss": 0.0081,
|
| 1262 |
+
"step": 89500
|
| 1263 |
+
},
|
| 1264 |
+
{
|
| 1265 |
+
"epoch": 123.45679012345678,
|
| 1266 |
+
"grad_norm": 0.050256237387657166,
|
| 1267 |
+
"learning_rate": 0.00038271604938271603,
|
| 1268 |
+
"loss": 0.0081,
|
| 1269 |
+
"step": 90000
|
| 1270 |
+
},
|
| 1271 |
+
{
|
| 1272 |
+
"epoch": 124.14266117969822,
|
| 1273 |
+
"grad_norm": 0.03073493391275406,
|
| 1274 |
+
"learning_rate": 0.0003792866941015089,
|
| 1275 |
+
"loss": 0.0076,
|
| 1276 |
+
"step": 90500
|
| 1277 |
+
},
|
| 1278 |
+
{
|
| 1279 |
+
"epoch": 124.82853223593965,
|
| 1280 |
+
"grad_norm": 0.07006064057350159,
|
| 1281 |
+
"learning_rate": 0.0003758573388203018,
|
| 1282 |
+
"loss": 0.0075,
|
| 1283 |
+
"step": 91000
|
| 1284 |
+
},
|
| 1285 |
+
{
|
| 1286 |
+
"epoch": 125.51440329218107,
|
| 1287 |
+
"grad_norm": 0.05567869916558266,
|
| 1288 |
+
"learning_rate": 0.00037242798353909467,
|
| 1289 |
+
"loss": 0.0079,
|
| 1290 |
+
"step": 91500
|
| 1291 |
+
},
|
| 1292 |
+
{
|
| 1293 |
+
"epoch": 126.2002743484225,
|
| 1294 |
+
"grad_norm": 0.03926622495055199,
|
| 1295 |
+
"learning_rate": 0.00036899862825788755,
|
| 1296 |
+
"loss": 0.008,
|
| 1297 |
+
"step": 92000
|
| 1298 |
+
},
|
| 1299 |
+
{
|
| 1300 |
+
"epoch": 126.88614540466392,
|
| 1301 |
+
"grad_norm": 0.05135864019393921,
|
| 1302 |
+
"learning_rate": 0.0003655692729766804,
|
| 1303 |
+
"loss": 0.0075,
|
| 1304 |
+
"step": 92500
|
| 1305 |
+
},
|
| 1306 |
+
{
|
| 1307 |
+
"epoch": 127.57201646090535,
|
| 1308 |
+
"grad_norm": 0.09010568261146545,
|
| 1309 |
+
"learning_rate": 0.00036213991769547325,
|
| 1310 |
+
"loss": 0.0068,
|
| 1311 |
+
"step": 93000
|
| 1312 |
+
},
|
| 1313 |
+
{
|
| 1314 |
+
"epoch": 128.25788751714677,
|
| 1315 |
+
"grad_norm": 0.043294016271829605,
|
| 1316 |
+
"learning_rate": 0.00035871056241426613,
|
| 1317 |
+
"loss": 0.0072,
|
| 1318 |
+
"step": 93500
|
| 1319 |
+
},
|
| 1320 |
+
{
|
| 1321 |
+
"epoch": 128.9437585733882,
|
| 1322 |
+
"grad_norm": 0.07277141511440277,
|
| 1323 |
+
"learning_rate": 0.000355281207133059,
|
| 1324 |
+
"loss": 0.0073,
|
| 1325 |
+
"step": 94000
|
| 1326 |
+
},
|
| 1327 |
+
{
|
| 1328 |
+
"epoch": 129.62962962962962,
|
| 1329 |
+
"grad_norm": 0.06138594448566437,
|
| 1330 |
+
"learning_rate": 0.0003518518518518519,
|
| 1331 |
+
"loss": 0.007,
|
| 1332 |
+
"step": 94500
|
| 1333 |
+
},
|
| 1334 |
+
{
|
| 1335 |
+
"epoch": 130.31550068587106,
|
| 1336 |
+
"grad_norm": 0.06205645576119423,
|
| 1337 |
+
"learning_rate": 0.0003484224965706447,
|
| 1338 |
+
"loss": 0.0078,
|
| 1339 |
+
"step": 95000
|
| 1340 |
+
},
|
| 1341 |
+
{
|
| 1342 |
+
"epoch": 131.0013717421125,
|
| 1343 |
+
"grad_norm": 0.0695485770702362,
|
| 1344 |
+
"learning_rate": 0.0003449931412894376,
|
| 1345 |
+
"loss": 0.0074,
|
| 1346 |
+
"step": 95500
|
| 1347 |
+
},
|
| 1348 |
+
{
|
| 1349 |
+
"epoch": 131.6872427983539,
|
| 1350 |
+
"grad_norm": 0.027826467528939247,
|
| 1351 |
+
"learning_rate": 0.0003415637860082305,
|
| 1352 |
+
"loss": 0.0063,
|
| 1353 |
+
"step": 96000
|
| 1354 |
+
},
|
| 1355 |
+
{
|
| 1356 |
+
"epoch": 132.37311385459535,
|
| 1357 |
+
"grad_norm": 0.05079466477036476,
|
| 1358 |
+
"learning_rate": 0.0003381344307270233,
|
| 1359 |
+
"loss": 0.0068,
|
| 1360 |
+
"step": 96500
|
| 1361 |
+
},
|
| 1362 |
+
{
|
| 1363 |
+
"epoch": 133.05898491083676,
|
| 1364 |
+
"grad_norm": 0.031004609540104866,
|
| 1365 |
+
"learning_rate": 0.0003347050754458162,
|
| 1366 |
+
"loss": 0.0076,
|
| 1367 |
+
"step": 97000
|
| 1368 |
+
},
|
| 1369 |
+
{
|
| 1370 |
+
"epoch": 133.7448559670782,
|
| 1371 |
+
"grad_norm": 0.04684276878833771,
|
| 1372 |
+
"learning_rate": 0.00033127572016460906,
|
| 1373 |
+
"loss": 0.0066,
|
| 1374 |
+
"step": 97500
|
| 1375 |
+
},
|
| 1376 |
+
{
|
| 1377 |
+
"epoch": 134.4307270233196,
|
| 1378 |
+
"grad_norm": 0.04819253832101822,
|
| 1379 |
+
"learning_rate": 0.00032784636488340194,
|
| 1380 |
+
"loss": 0.0078,
|
| 1381 |
+
"step": 98000
|
| 1382 |
+
},
|
| 1383 |
+
{
|
| 1384 |
+
"epoch": 135.11659807956104,
|
| 1385 |
+
"grad_norm": 0.04323802888393402,
|
| 1386 |
+
"learning_rate": 0.00032441700960219477,
|
| 1387 |
+
"loss": 0.0066,
|
| 1388 |
+
"step": 98500
|
| 1389 |
+
},
|
| 1390 |
+
{
|
| 1391 |
+
"epoch": 135.80246913580248,
|
| 1392 |
+
"grad_norm": 0.0720980316400528,
|
| 1393 |
+
"learning_rate": 0.00032098765432098765,
|
| 1394 |
+
"loss": 0.006,
|
| 1395 |
+
"step": 99000
|
| 1396 |
+
},
|
| 1397 |
+
{
|
| 1398 |
+
"epoch": 136.4883401920439,
|
| 1399 |
+
"grad_norm": 0.03386896476149559,
|
| 1400 |
+
"learning_rate": 0.0003175582990397805,
|
| 1401 |
+
"loss": 0.0061,
|
| 1402 |
+
"step": 99500
|
| 1403 |
+
},
|
| 1404 |
+
{
|
| 1405 |
+
"epoch": 137.17421124828533,
|
| 1406 |
+
"grad_norm": 0.033124495297670364,
|
| 1407 |
+
"learning_rate": 0.00031412894375857335,
|
| 1408 |
+
"loss": 0.0062,
|
| 1409 |
+
"step": 100000
|
| 1410 |
+
},
|
| 1411 |
+
{
|
| 1412 |
+
"epoch": 137.86008230452674,
|
| 1413 |
+
"grad_norm": 0.05887928605079651,
|
| 1414 |
+
"learning_rate": 0.00031069958847736623,
|
| 1415 |
+
"loss": 0.0061,
|
| 1416 |
+
"step": 100500
|
| 1417 |
+
},
|
| 1418 |
+
{
|
| 1419 |
+
"epoch": 138.54595336076818,
|
| 1420 |
+
"grad_norm": 0.04949059709906578,
|
| 1421 |
+
"learning_rate": 0.00030727023319615916,
|
| 1422 |
+
"loss": 0.0064,
|
| 1423 |
+
"step": 101000
|
| 1424 |
+
},
|
| 1425 |
+
{
|
| 1426 |
+
"epoch": 139.2318244170096,
|
| 1427 |
+
"grad_norm": 0.020787570625543594,
|
| 1428 |
+
"learning_rate": 0.00030384087791495204,
|
| 1429 |
+
"loss": 0.0057,
|
| 1430 |
+
"step": 101500
|
| 1431 |
+
},
|
| 1432 |
+
{
|
| 1433 |
+
"epoch": 139.91769547325103,
|
| 1434 |
+
"grad_norm": 0.02130681276321411,
|
| 1435 |
+
"learning_rate": 0.00030041152263374487,
|
| 1436 |
+
"loss": 0.0062,
|
| 1437 |
+
"step": 102000
|
| 1438 |
+
},
|
| 1439 |
+
{
|
| 1440 |
+
"epoch": 140.60356652949247,
|
| 1441 |
+
"grad_norm": 0.025823501870036125,
|
| 1442 |
+
"learning_rate": 0.00029698216735253775,
|
| 1443 |
+
"loss": 0.0062,
|
| 1444 |
+
"step": 102500
|
| 1445 |
+
},
|
| 1446 |
+
{
|
| 1447 |
+
"epoch": 141.28943758573388,
|
| 1448 |
+
"grad_norm": 0.06032751873135567,
|
| 1449 |
+
"learning_rate": 0.00029355281207133063,
|
| 1450 |
+
"loss": 0.0059,
|
| 1451 |
+
"step": 103000
|
| 1452 |
+
},
|
| 1453 |
+
{
|
| 1454 |
+
"epoch": 141.97530864197532,
|
| 1455 |
+
"grad_norm": 0.05369709059596062,
|
| 1456 |
+
"learning_rate": 0.00029012345679012345,
|
| 1457 |
+
"loss": 0.0067,
|
| 1458 |
+
"step": 103500
|
| 1459 |
+
},
|
| 1460 |
+
{
|
| 1461 |
+
"epoch": 142.66117969821673,
|
| 1462 |
+
"grad_norm": 0.012244037352502346,
|
| 1463 |
+
"learning_rate": 0.00028669410150891633,
|
| 1464 |
+
"loss": 0.0055,
|
| 1465 |
+
"step": 104000
|
| 1466 |
+
},
|
| 1467 |
+
{
|
| 1468 |
+
"epoch": 143.34705075445817,
|
| 1469 |
+
"grad_norm": 0.03182597830891609,
|
| 1470 |
+
"learning_rate": 0.0002832647462277092,
|
| 1471 |
+
"loss": 0.0057,
|
| 1472 |
+
"step": 104500
|
| 1473 |
+
},
|
| 1474 |
+
{
|
| 1475 |
+
"epoch": 144.03292181069958,
|
| 1476 |
+
"grad_norm": 0.11879345774650574,
|
| 1477 |
+
"learning_rate": 0.0002798353909465021,
|
| 1478 |
+
"loss": 0.0058,
|
| 1479 |
+
"step": 105000
|
| 1480 |
+
},
|
| 1481 |
+
{
|
| 1482 |
+
"epoch": 144.71879286694102,
|
| 1483 |
+
"grad_norm": 0.06132959946990013,
|
| 1484 |
+
"learning_rate": 0.0002764060356652949,
|
| 1485 |
+
"loss": 0.0059,
|
| 1486 |
+
"step": 105500
|
| 1487 |
+
},
|
| 1488 |
+
{
|
| 1489 |
+
"epoch": 145.40466392318245,
|
| 1490 |
+
"grad_norm": 0.03873404487967491,
|
| 1491 |
+
"learning_rate": 0.0002729766803840878,
|
| 1492 |
+
"loss": 0.0054,
|
| 1493 |
+
"step": 106000
|
| 1494 |
+
},
|
| 1495 |
+
{
|
| 1496 |
+
"epoch": 146.09053497942386,
|
| 1497 |
+
"grad_norm": 0.05853896215558052,
|
| 1498 |
+
"learning_rate": 0.0002695473251028807,
|
| 1499 |
+
"loss": 0.0055,
|
| 1500 |
+
"step": 106500
|
| 1501 |
+
},
|
| 1502 |
+
{
|
| 1503 |
+
"epoch": 146.7764060356653,
|
| 1504 |
+
"grad_norm": 0.022752197459340096,
|
| 1505 |
+
"learning_rate": 0.0002661179698216735,
|
| 1506 |
+
"loss": 0.0054,
|
| 1507 |
+
"step": 107000
|
| 1508 |
+
},
|
| 1509 |
+
{
|
| 1510 |
+
"epoch": 147.4622770919067,
|
| 1511 |
+
"grad_norm": 0.013148725032806396,
|
| 1512 |
+
"learning_rate": 0.0002626886145404664,
|
| 1513 |
+
"loss": 0.0052,
|
| 1514 |
+
"step": 107500
|
| 1515 |
+
},
|
| 1516 |
+
{
|
| 1517 |
+
"epoch": 148.14814814814815,
|
| 1518 |
+
"grad_norm": 0.0366031639277935,
|
| 1519 |
+
"learning_rate": 0.00025925925925925926,
|
| 1520 |
+
"loss": 0.0057,
|
| 1521 |
+
"step": 108000
|
| 1522 |
+
},
|
| 1523 |
+
{
|
| 1524 |
+
"epoch": 148.83401920438956,
|
| 1525 |
+
"grad_norm": 0.08972814679145813,
|
| 1526 |
+
"learning_rate": 0.0002558299039780521,
|
| 1527 |
+
"loss": 0.0055,
|
| 1528 |
+
"step": 108500
|
| 1529 |
+
},
|
| 1530 |
+
{
|
| 1531 |
+
"epoch": 149.519890260631,
|
| 1532 |
+
"grad_norm": 0.034281570464372635,
|
| 1533 |
+
"learning_rate": 0.00025240054869684497,
|
| 1534 |
+
"loss": 0.0051,
|
| 1535 |
+
"step": 109000
|
| 1536 |
+
},
|
| 1537 |
+
{
|
| 1538 |
+
"epoch": 150.20576131687244,
|
| 1539 |
+
"grad_norm": 0.010903775691986084,
|
| 1540 |
+
"learning_rate": 0.00024897119341563785,
|
| 1541 |
+
"loss": 0.005,
|
| 1542 |
+
"step": 109500
|
| 1543 |
+
},
|
| 1544 |
+
{
|
| 1545 |
+
"epoch": 150.89163237311385,
|
| 1546 |
+
"grad_norm": 0.015872234478592873,
|
| 1547 |
+
"learning_rate": 0.0002455418381344307,
|
| 1548 |
+
"loss": 0.0048,
|
| 1549 |
+
"step": 110000
|
| 1550 |
+
},
|
| 1551 |
+
{
|
| 1552 |
+
"epoch": 151.5775034293553,
|
| 1553 |
+
"grad_norm": 0.03331177309155464,
|
| 1554 |
+
"learning_rate": 0.0002421124828532236,
|
| 1555 |
+
"loss": 0.0053,
|
| 1556 |
+
"step": 110500
|
| 1557 |
+
},
|
| 1558 |
+
{
|
| 1559 |
+
"epoch": 152.2633744855967,
|
| 1560 |
+
"grad_norm": 0.018091242760419846,
|
| 1561 |
+
"learning_rate": 0.00023868312757201646,
|
| 1562 |
+
"loss": 0.0052,
|
| 1563 |
+
"step": 111000
|
| 1564 |
+
},
|
| 1565 |
+
{
|
| 1566 |
+
"epoch": 152.94924554183814,
|
| 1567 |
+
"grad_norm": 0.03916322439908981,
|
| 1568 |
+
"learning_rate": 0.00023525377229080934,
|
| 1569 |
+
"loss": 0.0051,
|
| 1570 |
+
"step": 111500
|
| 1571 |
+
},
|
| 1572 |
+
{
|
| 1573 |
+
"epoch": 153.63511659807955,
|
| 1574 |
+
"grad_norm": 0.023662962019443512,
|
| 1575 |
+
"learning_rate": 0.0002318244170096022,
|
| 1576 |
+
"loss": 0.0048,
|
| 1577 |
+
"step": 112000
|
| 1578 |
+
},
|
| 1579 |
+
{
|
| 1580 |
+
"epoch": 154.320987654321,
|
| 1581 |
+
"grad_norm": 0.014827713370323181,
|
| 1582 |
+
"learning_rate": 0.00022839506172839504,
|
| 1583 |
+
"loss": 0.0046,
|
| 1584 |
+
"step": 112500
|
| 1585 |
+
},
|
| 1586 |
+
{
|
| 1587 |
+
"epoch": 155.00685871056243,
|
| 1588 |
+
"grad_norm": 0.03849729895591736,
|
| 1589 |
+
"learning_rate": 0.00022496570644718795,
|
| 1590 |
+
"loss": 0.0047,
|
| 1591 |
+
"step": 113000
|
| 1592 |
+
},
|
| 1593 |
+
{
|
| 1594 |
+
"epoch": 155.69272976680384,
|
| 1595 |
+
"grad_norm": 0.057328708469867706,
|
| 1596 |
+
"learning_rate": 0.0002215363511659808,
|
| 1597 |
+
"loss": 0.0046,
|
| 1598 |
+
"step": 113500
|
| 1599 |
+
},
|
| 1600 |
+
{
|
| 1601 |
+
"epoch": 156.37860082304528,
|
| 1602 |
+
"grad_norm": 0.017304692417383194,
|
| 1603 |
+
"learning_rate": 0.00021810699588477368,
|
| 1604 |
+
"loss": 0.0047,
|
| 1605 |
+
"step": 114000
|
| 1606 |
+
},
|
| 1607 |
+
{
|
| 1608 |
+
"epoch": 157.06447187928669,
|
| 1609 |
+
"grad_norm": 0.02947048842906952,
|
| 1610 |
+
"learning_rate": 0.00021467764060356654,
|
| 1611 |
+
"loss": 0.0047,
|
| 1612 |
+
"step": 114500
|
| 1613 |
+
},
|
| 1614 |
+
{
|
| 1615 |
+
"epoch": 157.75034293552812,
|
| 1616 |
+
"grad_norm": 0.02040564827620983,
|
| 1617 |
+
"learning_rate": 0.0002112482853223594,
|
| 1618 |
+
"loss": 0.0048,
|
| 1619 |
+
"step": 115000
|
| 1620 |
+
},
|
| 1621 |
+
{
|
| 1622 |
+
"epoch": 158.43621399176953,
|
| 1623 |
+
"grad_norm": 0.05288396403193474,
|
| 1624 |
+
"learning_rate": 0.00020781893004115227,
|
| 1625 |
+
"loss": 0.0047,
|
| 1626 |
+
"step": 115500
|
| 1627 |
+
},
|
| 1628 |
+
{
|
| 1629 |
+
"epoch": 159.12208504801097,
|
| 1630 |
+
"grad_norm": 0.026866160333156586,
|
| 1631 |
+
"learning_rate": 0.00020438957475994512,
|
| 1632 |
+
"loss": 0.0047,
|
| 1633 |
+
"step": 116000
|
| 1634 |
+
},
|
| 1635 |
+
{
|
| 1636 |
+
"epoch": 159.8079561042524,
|
| 1637 |
+
"grad_norm": 0.04552701115608215,
|
| 1638 |
+
"learning_rate": 0.000200960219478738,
|
| 1639 |
+
"loss": 0.0043,
|
| 1640 |
+
"step": 116500
|
| 1641 |
+
},
|
| 1642 |
+
{
|
| 1643 |
+
"epoch": 160.49382716049382,
|
| 1644 |
+
"grad_norm": 0.020935742184519768,
|
| 1645 |
+
"learning_rate": 0.00019753086419753085,
|
| 1646 |
+
"loss": 0.0045,
|
| 1647 |
+
"step": 117000
|
| 1648 |
+
},
|
| 1649 |
+
{
|
| 1650 |
+
"epoch": 161.17969821673526,
|
| 1651 |
+
"grad_norm": 0.03535682335495949,
|
| 1652 |
+
"learning_rate": 0.00019410150891632373,
|
| 1653 |
+
"loss": 0.0044,
|
| 1654 |
+
"step": 117500
|
| 1655 |
+
},
|
| 1656 |
+
{
|
| 1657 |
+
"epoch": 161.86556927297667,
|
| 1658 |
+
"grad_norm": 0.019771935418248177,
|
| 1659 |
+
"learning_rate": 0.0001906721536351166,
|
| 1660 |
+
"loss": 0.0044,
|
| 1661 |
+
"step": 118000
|
| 1662 |
+
},
|
| 1663 |
+
{
|
| 1664 |
+
"epoch": 162.5514403292181,
|
| 1665 |
+
"grad_norm": 0.013906066305935383,
|
| 1666 |
+
"learning_rate": 0.00018724279835390946,
|
| 1667 |
+
"loss": 0.004,
|
| 1668 |
+
"step": 118500
|
| 1669 |
+
},
|
| 1670 |
+
{
|
| 1671 |
+
"epoch": 163.23731138545952,
|
| 1672 |
+
"grad_norm": 0.012126780115067959,
|
| 1673 |
+
"learning_rate": 0.00018381344307270234,
|
| 1674 |
+
"loss": 0.0043,
|
| 1675 |
+
"step": 119000
|
| 1676 |
+
},
|
| 1677 |
+
{
|
| 1678 |
+
"epoch": 163.92318244170096,
|
| 1679 |
+
"grad_norm": 0.021324895322322845,
|
| 1680 |
+
"learning_rate": 0.0001803840877914952,
|
| 1681 |
+
"loss": 0.0043,
|
| 1682 |
+
"step": 119500
|
| 1683 |
+
},
|
| 1684 |
+
{
|
| 1685 |
+
"epoch": 164.6090534979424,
|
| 1686 |
+
"grad_norm": 0.013377720490098,
|
| 1687 |
+
"learning_rate": 0.00017695473251028808,
|
| 1688 |
+
"loss": 0.0043,
|
| 1689 |
+
"step": 120000
|
| 1690 |
+
},
|
| 1691 |
+
{
|
| 1692 |
+
"epoch": 165.2949245541838,
|
| 1693 |
+
"grad_norm": 0.08576662838459015,
|
| 1694 |
+
"learning_rate": 0.00017352537722908093,
|
| 1695 |
+
"loss": 0.0043,
|
| 1696 |
+
"step": 120500
|
| 1697 |
+
},
|
| 1698 |
+
{
|
| 1699 |
+
"epoch": 165.98079561042525,
|
| 1700 |
+
"grad_norm": 0.008204947225749493,
|
| 1701 |
+
"learning_rate": 0.0001700960219478738,
|
| 1702 |
+
"loss": 0.004,
|
| 1703 |
+
"step": 121000
|
| 1704 |
+
},
|
| 1705 |
+
{
|
| 1706 |
+
"epoch": 166.66666666666666,
|
| 1707 |
+
"grad_norm": 0.01228815782815218,
|
| 1708 |
+
"learning_rate": 0.00016666666666666666,
|
| 1709 |
+
"loss": 0.0038,
|
| 1710 |
+
"step": 121500
|
| 1711 |
+
},
|
| 1712 |
+
{
|
| 1713 |
+
"epoch": 167.3525377229081,
|
| 1714 |
+
"grad_norm": 0.007976936176419258,
|
| 1715 |
+
"learning_rate": 0.00016323731138545951,
|
| 1716 |
+
"loss": 0.004,
|
| 1717 |
+
"step": 122000
|
| 1718 |
+
},
|
| 1719 |
+
{
|
| 1720 |
+
"epoch": 168.0384087791495,
|
| 1721 |
+
"grad_norm": 0.06643296033143997,
|
| 1722 |
+
"learning_rate": 0.00015980795610425242,
|
| 1723 |
+
"loss": 0.0042,
|
| 1724 |
+
"step": 122500
|
| 1725 |
+
},
|
| 1726 |
+
{
|
| 1727 |
+
"epoch": 168.72427983539094,
|
| 1728 |
+
"grad_norm": 0.040761686861515045,
|
| 1729 |
+
"learning_rate": 0.00015637860082304527,
|
| 1730 |
+
"loss": 0.0041,
|
| 1731 |
+
"step": 123000
|
| 1732 |
+
},
|
| 1733 |
+
{
|
| 1734 |
+
"epoch": 169.41015089163238,
|
| 1735 |
+
"grad_norm": 0.006533100735396147,
|
| 1736 |
+
"learning_rate": 0.00015294924554183815,
|
| 1737 |
+
"loss": 0.0039,
|
| 1738 |
+
"step": 123500
|
| 1739 |
+
},
|
| 1740 |
+
{
|
| 1741 |
+
"epoch": 170.0960219478738,
|
| 1742 |
+
"grad_norm": 0.015169362537562847,
|
| 1743 |
+
"learning_rate": 0.000149519890260631,
|
| 1744 |
+
"loss": 0.0039,
|
| 1745 |
+
"step": 124000
|
| 1746 |
+
},
|
| 1747 |
+
{
|
| 1748 |
+
"epoch": 170.78189300411523,
|
| 1749 |
+
"grad_norm": 0.03795445337891579,
|
| 1750 |
+
"learning_rate": 0.00014609053497942388,
|
| 1751 |
+
"loss": 0.0039,
|
| 1752 |
+
"step": 124500
|
| 1753 |
+
},
|
| 1754 |
+
{
|
| 1755 |
+
"epoch": 171.46776406035664,
|
| 1756 |
+
"grad_norm": 0.03991511091589928,
|
| 1757 |
+
"learning_rate": 0.00014266117969821674,
|
| 1758 |
+
"loss": 0.0038,
|
| 1759 |
+
"step": 125000
|
| 1760 |
+
},
|
| 1761 |
+
{
|
| 1762 |
+
"epoch": 172.15363511659808,
|
| 1763 |
+
"grad_norm": 0.008595280349254608,
|
| 1764 |
+
"learning_rate": 0.0001392318244170096,
|
| 1765 |
+
"loss": 0.0039,
|
| 1766 |
+
"step": 125500
|
| 1767 |
+
},
|
| 1768 |
+
{
|
| 1769 |
+
"epoch": 172.8395061728395,
|
| 1770 |
+
"grad_norm": 0.01920117437839508,
|
| 1771 |
+
"learning_rate": 0.00013580246913580247,
|
| 1772 |
+
"loss": 0.0036,
|
| 1773 |
+
"step": 126000
|
| 1774 |
+
},
|
| 1775 |
+
{
|
| 1776 |
+
"epoch": 173.52537722908093,
|
| 1777 |
+
"grad_norm": 0.00689229741692543,
|
| 1778 |
+
"learning_rate": 0.00013237311385459532,
|
| 1779 |
+
"loss": 0.0035,
|
| 1780 |
+
"step": 126500
|
| 1781 |
+
},
|
| 1782 |
+
{
|
| 1783 |
+
"epoch": 174.21124828532237,
|
| 1784 |
+
"grad_norm": 0.007589759770780802,
|
| 1785 |
+
"learning_rate": 0.0001289437585733882,
|
| 1786 |
+
"loss": 0.0037,
|
| 1787 |
+
"step": 127000
|
| 1788 |
+
},
|
| 1789 |
+
{
|
| 1790 |
+
"epoch": 174.89711934156378,
|
| 1791 |
+
"grad_norm": 0.026523003354668617,
|
| 1792 |
+
"learning_rate": 0.00012551440329218108,
|
| 1793 |
+
"loss": 0.0037,
|
| 1794 |
+
"step": 127500
|
| 1795 |
+
},
|
| 1796 |
+
{
|
| 1797 |
+
"epoch": 175.58299039780522,
|
| 1798 |
+
"grad_norm": 0.009044776670634747,
|
| 1799 |
+
"learning_rate": 0.00012208504801097393,
|
| 1800 |
+
"loss": 0.0035,
|
| 1801 |
+
"step": 128000
|
| 1802 |
+
},
|
| 1803 |
+
{
|
| 1804 |
+
"epoch": 176.26886145404663,
|
| 1805 |
+
"grad_norm": 0.007540062535554171,
|
| 1806 |
+
"learning_rate": 0.00011865569272976681,
|
| 1807 |
+
"loss": 0.0038,
|
| 1808 |
+
"step": 128500
|
| 1809 |
+
},
|
| 1810 |
+
{
|
| 1811 |
+
"epoch": 176.95473251028807,
|
| 1812 |
+
"grad_norm": 0.00613565556704998,
|
| 1813 |
+
"learning_rate": 0.00011522633744855968,
|
| 1814 |
+
"loss": 0.0036,
|
| 1815 |
+
"step": 129000
|
| 1816 |
+
},
|
| 1817 |
+
{
|
| 1818 |
+
"epoch": 177.6406035665295,
|
| 1819 |
+
"grad_norm": 0.005932086147367954,
|
| 1820 |
+
"learning_rate": 0.00011179698216735255,
|
| 1821 |
+
"loss": 0.0036,
|
| 1822 |
+
"step": 129500
|
| 1823 |
+
},
|
| 1824 |
+
{
|
| 1825 |
+
"epoch": 178.32647462277092,
|
| 1826 |
+
"grad_norm": 0.0038564407732337713,
|
| 1827 |
+
"learning_rate": 0.0001083676268861454,
|
| 1828 |
+
"loss": 0.0035,
|
| 1829 |
+
"step": 130000
|
| 1830 |
+
},
|
| 1831 |
+
{
|
| 1832 |
+
"epoch": 179.01234567901236,
|
| 1833 |
+
"grad_norm": 0.10642554610967636,
|
| 1834 |
+
"learning_rate": 0.00010493827160493826,
|
| 1835 |
+
"loss": 0.0035,
|
| 1836 |
+
"step": 130500
|
| 1837 |
+
},
|
| 1838 |
+
{
|
| 1839 |
+
"epoch": 179.69821673525377,
|
| 1840 |
+
"grad_norm": 0.010230828076601028,
|
| 1841 |
+
"learning_rate": 0.00010150891632373114,
|
| 1842 |
+
"loss": 0.0034,
|
| 1843 |
+
"step": 131000
|
| 1844 |
+
},
|
| 1845 |
+
{
|
| 1846 |
+
"epoch": 180.3840877914952,
|
| 1847 |
+
"grad_norm": 0.02526956982910633,
|
| 1848 |
+
"learning_rate": 9.807956104252401e-05,
|
| 1849 |
+
"loss": 0.0034,
|
| 1850 |
+
"step": 131500
|
| 1851 |
+
},
|
| 1852 |
+
{
|
| 1853 |
+
"epoch": 181.06995884773661,
|
| 1854 |
+
"grad_norm": 0.00888708233833313,
|
| 1855 |
+
"learning_rate": 9.465020576131688e-05,
|
| 1856 |
+
"loss": 0.0035,
|
| 1857 |
+
"step": 132000
|
| 1858 |
+
},
|
| 1859 |
+
{
|
| 1860 |
+
"epoch": 181.75582990397805,
|
| 1861 |
+
"grad_norm": 0.005969726946204901,
|
| 1862 |
+
"learning_rate": 9.122085048010974e-05,
|
| 1863 |
+
"loss": 0.0033,
|
| 1864 |
+
"step": 132500
|
| 1865 |
+
},
|
| 1866 |
+
{
|
| 1867 |
+
"epoch": 182.4417009602195,
|
| 1868 |
+
"grad_norm": 0.004369193222373724,
|
| 1869 |
+
"learning_rate": 8.77914951989026e-05,
|
| 1870 |
+
"loss": 0.0033,
|
| 1871 |
+
"step": 133000
|
| 1872 |
+
},
|
| 1873 |
+
{
|
| 1874 |
+
"epoch": 183.1275720164609,
|
| 1875 |
+
"grad_norm": 0.03928361088037491,
|
| 1876 |
+
"learning_rate": 8.436213991769547e-05,
|
| 1877 |
+
"loss": 0.0035,
|
| 1878 |
+
"step": 133500
|
| 1879 |
+
},
|
| 1880 |
+
{
|
| 1881 |
+
"epoch": 183.81344307270234,
|
| 1882 |
+
"grad_norm": 0.008007598109543324,
|
| 1883 |
+
"learning_rate": 8.093278463648834e-05,
|
| 1884 |
+
"loss": 0.0032,
|
| 1885 |
+
"step": 134000
|
| 1886 |
+
},
|
| 1887 |
+
{
|
| 1888 |
+
"epoch": 184.49931412894375,
|
| 1889 |
+
"grad_norm": 0.00541963754221797,
|
| 1890 |
+
"learning_rate": 7.750342935528121e-05,
|
| 1891 |
+
"loss": 0.0032,
|
| 1892 |
+
"step": 134500
|
| 1893 |
+
},
|
| 1894 |
+
{
|
| 1895 |
+
"epoch": 185.1851851851852,
|
| 1896 |
+
"grad_norm": 0.010783227160573006,
|
| 1897 |
+
"learning_rate": 7.407407407407407e-05,
|
| 1898 |
+
"loss": 0.0033,
|
| 1899 |
+
"step": 135000
|
| 1900 |
+
},
|
| 1901 |
+
{
|
| 1902 |
+
"epoch": 185.8710562414266,
|
| 1903 |
+
"grad_norm": 0.005976933054625988,
|
| 1904 |
+
"learning_rate": 7.064471879286695e-05,
|
| 1905 |
+
"loss": 0.0033,
|
| 1906 |
+
"step": 135500
|
| 1907 |
+
},
|
| 1908 |
+
{
|
| 1909 |
+
"epoch": 186.55692729766804,
|
| 1910 |
+
"grad_norm": 0.010424752719700336,
|
| 1911 |
+
"learning_rate": 6.721536351165982e-05,
|
| 1912 |
+
"loss": 0.0031,
|
| 1913 |
+
"step": 136000
|
| 1914 |
+
},
|
| 1915 |
+
{
|
| 1916 |
+
"epoch": 187.24279835390948,
|
| 1917 |
+
"grad_norm": 0.00834636203944683,
|
| 1918 |
+
"learning_rate": 6.378600823045267e-05,
|
| 1919 |
+
"loss": 0.0032,
|
| 1920 |
+
"step": 136500
|
| 1921 |
+
},
|
| 1922 |
+
{
|
| 1923 |
+
"epoch": 187.9286694101509,
|
| 1924 |
+
"grad_norm": 0.0064740655943751335,
|
| 1925 |
+
"learning_rate": 6.0356652949245544e-05,
|
| 1926 |
+
"loss": 0.0032,
|
| 1927 |
+
"step": 137000
|
| 1928 |
+
},
|
| 1929 |
+
{
|
| 1930 |
+
"epoch": 188.61454046639233,
|
| 1931 |
+
"grad_norm": 0.007856756448745728,
|
| 1932 |
+
"learning_rate": 5.692729766803841e-05,
|
| 1933 |
+
"loss": 0.0031,
|
| 1934 |
+
"step": 137500
|
| 1935 |
+
},
|
| 1936 |
+
{
|
| 1937 |
+
"epoch": 189.30041152263374,
|
| 1938 |
+
"grad_norm": 0.006533577106893063,
|
| 1939 |
+
"learning_rate": 5.3497942386831277e-05,
|
| 1940 |
+
"loss": 0.0031,
|
| 1941 |
+
"step": 138000
|
| 1942 |
+
},
|
| 1943 |
+
{
|
| 1944 |
+
"epoch": 189.98628257887518,
|
| 1945 |
+
"grad_norm": 0.005114428699016571,
|
| 1946 |
+
"learning_rate": 5.006858710562415e-05,
|
| 1947 |
+
"loss": 0.0032,
|
| 1948 |
+
"step": 138500
|
| 1949 |
+
},
|
| 1950 |
+
{
|
| 1951 |
+
"epoch": 190.6721536351166,
|
| 1952 |
+
"grad_norm": 0.00717920670285821,
|
| 1953 |
+
"learning_rate": 4.663923182441701e-05,
|
| 1954 |
+
"loss": 0.0031,
|
| 1955 |
+
"step": 139000
|
| 1956 |
+
},
|
| 1957 |
+
{
|
| 1958 |
+
"epoch": 191.35802469135803,
|
| 1959 |
+
"grad_norm": 0.008158246986567974,
|
| 1960 |
+
"learning_rate": 4.3209876543209875e-05,
|
| 1961 |
+
"loss": 0.0031,
|
| 1962 |
+
"step": 139500
|
| 1963 |
+
},
|
| 1964 |
+
{
|
| 1965 |
+
"epoch": 192.04389574759946,
|
| 1966 |
+
"grad_norm": 0.005865829065442085,
|
| 1967 |
+
"learning_rate": 3.978052126200275e-05,
|
| 1968 |
+
"loss": 0.0031,
|
| 1969 |
+
"step": 140000
|
| 1970 |
+
},
|
| 1971 |
+
{
|
| 1972 |
+
"epoch": 192.72976680384087,
|
| 1973 |
+
"grad_norm": 0.0036265423987060785,
|
| 1974 |
+
"learning_rate": 3.635116598079561e-05,
|
| 1975 |
+
"loss": 0.003,
|
| 1976 |
+
"step": 140500
|
| 1977 |
+
},
|
| 1978 |
+
{
|
| 1979 |
+
"epoch": 193.4156378600823,
|
| 1980 |
+
"grad_norm": 0.005760482046753168,
|
| 1981 |
+
"learning_rate": 3.292181069958848e-05,
|
| 1982 |
+
"loss": 0.0029,
|
| 1983 |
+
"step": 141000
|
| 1984 |
+
},
|
| 1985 |
+
{
|
| 1986 |
+
"epoch": 194.10150891632372,
|
| 1987 |
+
"grad_norm": 0.00619177520275116,
|
| 1988 |
+
"learning_rate": 2.9492455418381346e-05,
|
| 1989 |
+
"loss": 0.0031,
|
| 1990 |
+
"step": 141500
|
| 1991 |
+
},
|
| 1992 |
+
{
|
| 1993 |
+
"epoch": 194.78737997256516,
|
| 1994 |
+
"grad_norm": 0.00745520880445838,
|
| 1995 |
+
"learning_rate": 2.6063100137174212e-05,
|
| 1996 |
+
"loss": 0.0029,
|
| 1997 |
+
"step": 142000
|
| 1998 |
+
},
|
| 1999 |
+
{
|
| 2000 |
+
"epoch": 195.47325102880657,
|
| 2001 |
+
"grad_norm": 0.006654828321188688,
|
| 2002 |
+
"learning_rate": 2.2633744855967078e-05,
|
| 2003 |
+
"loss": 0.003,
|
| 2004 |
+
"step": 142500
|
| 2005 |
+
},
|
| 2006 |
+
{
|
| 2007 |
+
"epoch": 196.159122085048,
|
| 2008 |
+
"grad_norm": 0.007228308357298374,
|
| 2009 |
+
"learning_rate": 1.9204389574759948e-05,
|
| 2010 |
+
"loss": 0.003,
|
| 2011 |
+
"step": 143000
|
| 2012 |
+
},
|
| 2013 |
+
{
|
| 2014 |
+
"epoch": 196.84499314128945,
|
| 2015 |
+
"grad_norm": 0.007839037105441093,
|
| 2016 |
+
"learning_rate": 1.577503429355281e-05,
|
| 2017 |
+
"loss": 0.0029,
|
| 2018 |
+
"step": 143500
|
| 2019 |
+
},
|
| 2020 |
+
{
|
| 2021 |
+
"epoch": 197.53086419753086,
|
| 2022 |
+
"grad_norm": 0.005645914003252983,
|
| 2023 |
+
"learning_rate": 1.2345679012345678e-05,
|
| 2024 |
+
"loss": 0.0029,
|
| 2025 |
+
"step": 144000
|
| 2026 |
+
},
|
| 2027 |
+
{
|
| 2028 |
+
"epoch": 198.2167352537723,
|
| 2029 |
+
"grad_norm": 0.007110149599611759,
|
| 2030 |
+
"learning_rate": 8.916323731138546e-06,
|
| 2031 |
+
"loss": 0.0029,
|
| 2032 |
+
"step": 144500
|
| 2033 |
+
},
|
| 2034 |
+
{
|
| 2035 |
+
"epoch": 198.9026063100137,
|
| 2036 |
+
"grad_norm": 0.007023398298770189,
|
| 2037 |
+
"learning_rate": 5.486968449931412e-06,
|
| 2038 |
+
"loss": 0.003,
|
| 2039 |
+
"step": 145000
|
| 2040 |
+
},
|
| 2041 |
+
{
|
| 2042 |
+
"epoch": 199.58847736625515,
|
| 2043 |
+
"grad_norm": 0.006341323722153902,
|
| 2044 |
+
"learning_rate": 2.05761316872428e-06,
|
| 2045 |
+
"loss": 0.0028,
|
| 2046 |
+
"step": 145500
|
| 2047 |
+
},
|
| 2048 |
+
{
|
| 2049 |
+
"epoch": 200.0,
|
| 2050 |
+
"step": 145800,
|
| 2051 |
+
"total_flos": 2.53654993993728e+18,
|
| 2052 |
+
"train_loss": 0.006510871606680919,
|
| 2053 |
+
"train_runtime": 172601.0106,
|
| 2054 |
+
"train_samples_per_second": 3.377,
|
| 2055 |
+
"train_steps_per_second": 0.845
|
| 2056 |
+
}
|
| 2057 |
+
],
|
| 2058 |
+
"logging_steps": 500,
|
| 2059 |
+
"max_steps": 145800,
|
| 2060 |
+
"num_input_tokens_seen": 0,
|
| 2061 |
+
"num_train_epochs": 200,
|
| 2062 |
+
"save_steps": 500,
|
| 2063 |
+
"stateful_callbacks": {
|
| 2064 |
+
"TrainerControl": {
|
| 2065 |
+
"args": {
|
| 2066 |
+
"should_epoch_stop": false,
|
| 2067 |
+
"should_evaluate": false,
|
| 2068 |
+
"should_log": false,
|
| 2069 |
+
"should_save": true,
|
| 2070 |
+
"should_training_stop": true
|
| 2071 |
+
},
|
| 2072 |
+
"attributes": {}
|
| 2073 |
+
}
|
| 2074 |
+
},
|
| 2075 |
+
"total_flos": 2.53654993993728e+18,
|
| 2076 |
+
"train_batch_size": 4,
|
| 2077 |
+
"trial_name": null,
|
| 2078 |
+
"trial_params": null
|
| 2079 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ea0b9d37ab99617fb577d6314c306e902a72ff6b8364ab4827bd894874a27b5d
|
| 3 |
+
size 4667
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|