Upload 13 files
Browse files- added_tokens.json +3 -0
- all_results.json +8 -0
- config.json +34 -0
- generation_config.json +6 -0
- merges.txt +0 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +23 -0
- tokenizer.json +0 -0
- tokenizer_config.json +33 -0
- train_results.json +8 -0
- trainer_state.json +1039 -0
- training_args.bin +3 -0
- vocab.json +0 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|endoftext|>": 50257
|
| 3 |
+
}
|
all_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 2.6318450465295498,
|
| 4 |
+
"train_runtime": 20073.5525,
|
| 5 |
+
"train_samples": 6758,
|
| 6 |
+
"train_samples_per_second": 0.337,
|
| 7 |
+
"train_steps_per_second": 0.084
|
| 8 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,34 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/zstorage/AI/text-generation-webui/models/l_warhammer2",
|
| 3 |
+
"activation_function": "gelu_new",
|
| 4 |
+
"architectures": [
|
| 5 |
+
"GPT2LMHeadModel"
|
| 6 |
+
],
|
| 7 |
+
"attn_pdrop": 0.1,
|
| 8 |
+
"bos_token_id": 50256,
|
| 9 |
+
"embd_pdrop": 0.1,
|
| 10 |
+
"eos_token_id": 50256,
|
| 11 |
+
"gradient_checkpointing": false,
|
| 12 |
+
"initializer_range": 0.02,
|
| 13 |
+
"layer_norm_epsilon": 1e-05,
|
| 14 |
+
"model_type": "gpt2",
|
| 15 |
+
"n_ctx": 2048,
|
| 16 |
+
"n_embd": 1536,
|
| 17 |
+
"n_head": 16,
|
| 18 |
+
"n_inner": null,
|
| 19 |
+
"n_layer": 24,
|
| 20 |
+
"n_positions": 2048,
|
| 21 |
+
"reorder_and_upcast_attn": false,
|
| 22 |
+
"resid_pdrop": 0.1,
|
| 23 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 24 |
+
"scale_attn_weights": true,
|
| 25 |
+
"summary_activation": null,
|
| 26 |
+
"summary_first_dropout": 0.1,
|
| 27 |
+
"summary_proj_to_labels": true,
|
| 28 |
+
"summary_type": "cls_index",
|
| 29 |
+
"summary_use_proj": true,
|
| 30 |
+
"torch_dtype": "float32",
|
| 31 |
+
"transformers_version": "4.30.0.dev0",
|
| 32 |
+
"use_cache": true,
|
| 33 |
+
"vocab_size": 50258
|
| 34 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"eos_token_id": 50256,
|
| 5 |
+
"transformers_version": "4.30.0.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 sha256:744f1b05194cd2ddc74a55cef099e2581430677a9774235aa161ee823c1360e0
|
| 3 |
+
size 3041301910
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,23 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": {
|
| 3 |
+
"content": "<|endoftext|>",
|
| 4 |
+
"lstrip": false,
|
| 5 |
+
"normalized": true,
|
| 6 |
+
"rstrip": false,
|
| 7 |
+
"single_word": false
|
| 8 |
+
},
|
| 9 |
+
"eos_token": {
|
| 10 |
+
"content": "<|endoftext|>",
|
| 11 |
+
"lstrip": false,
|
| 12 |
+
"normalized": true,
|
| 13 |
+
"rstrip": false,
|
| 14 |
+
"single_word": false
|
| 15 |
+
},
|
| 16 |
+
"unk_token": {
|
| 17 |
+
"content": "<|endoftext|>",
|
| 18 |
+
"lstrip": false,
|
| 19 |
+
"normalized": true,
|
| 20 |
+
"rstrip": false,
|
| 21 |
+
"single_word": false
|
| 22 |
+
}
|
| 23 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,33 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_bos_token": false,
|
| 3 |
+
"add_prefix_space": false,
|
| 4 |
+
"bos_token": {
|
| 5 |
+
"__type": "AddedToken",
|
| 6 |
+
"content": "<|endoftext|>",
|
| 7 |
+
"lstrip": false,
|
| 8 |
+
"normalized": true,
|
| 9 |
+
"rstrip": false,
|
| 10 |
+
"single_word": false
|
| 11 |
+
},
|
| 12 |
+
"clean_up_tokenization_spaces": true,
|
| 13 |
+
"eos_token": {
|
| 14 |
+
"__type": "AddedToken",
|
| 15 |
+
"content": "<|endoftext|>",
|
| 16 |
+
"lstrip": false,
|
| 17 |
+
"normalized": true,
|
| 18 |
+
"rstrip": false,
|
| 19 |
+
"single_word": false
|
| 20 |
+
},
|
| 21 |
+
"errors": "replace",
|
| 22 |
+
"model_max_length": 1000000000000000019884624838656,
|
| 23 |
+
"pad_token": null,
|
| 24 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 25 |
+
"unk_token": {
|
| 26 |
+
"__type": "AddedToken",
|
| 27 |
+
"content": "<|endoftext|>",
|
| 28 |
+
"lstrip": false,
|
| 29 |
+
"normalized": true,
|
| 30 |
+
"rstrip": false,
|
| 31 |
+
"single_word": false
|
| 32 |
+
}
|
| 33 |
+
}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 1.0,
|
| 3 |
+
"train_loss": 2.6318450465295498,
|
| 4 |
+
"train_runtime": 20073.5525,
|
| 5 |
+
"train_samples": 6758,
|
| 6 |
+
"train_samples_per_second": 0.337,
|
| 7 |
+
"train_steps_per_second": 0.084
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,1039 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9997040544539805,
|
| 5 |
+
"global_step": 1689,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.0,
|
| 12 |
+
"learning_rate": 5.000000000000001e-07,
|
| 13 |
+
"loss": 2.7237,
|
| 14 |
+
"step": 1
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.01,
|
| 18 |
+
"learning_rate": 5e-06,
|
| 19 |
+
"loss": 2.7145,
|
| 20 |
+
"step": 10
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.01,
|
| 24 |
+
"learning_rate": 1e-05,
|
| 25 |
+
"loss": 2.7579,
|
| 26 |
+
"step": 20
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.02,
|
| 30 |
+
"learning_rate": 1.5e-05,
|
| 31 |
+
"loss": 2.747,
|
| 32 |
+
"step": 30
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.02,
|
| 36 |
+
"learning_rate": 2e-05,
|
| 37 |
+
"loss": 2.6339,
|
| 38 |
+
"step": 40
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.03,
|
| 42 |
+
"learning_rate": 2.5e-05,
|
| 43 |
+
"loss": 2.6645,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.04,
|
| 48 |
+
"learning_rate": 3e-05,
|
| 49 |
+
"loss": 2.6576,
|
| 50 |
+
"step": 60
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.04,
|
| 54 |
+
"learning_rate": 3.5e-05,
|
| 55 |
+
"loss": 2.6678,
|
| 56 |
+
"step": 70
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.05,
|
| 60 |
+
"learning_rate": 4e-05,
|
| 61 |
+
"loss": 2.5723,
|
| 62 |
+
"step": 80
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.05,
|
| 66 |
+
"learning_rate": 4.5e-05,
|
| 67 |
+
"loss": 2.514,
|
| 68 |
+
"step": 90
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.06,
|
| 72 |
+
"learning_rate": 5e-05,
|
| 73 |
+
"loss": 2.575,
|
| 74 |
+
"step": 100
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.07,
|
| 78 |
+
"learning_rate": 4.968533668974198e-05,
|
| 79 |
+
"loss": 2.5101,
|
| 80 |
+
"step": 110
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.07,
|
| 84 |
+
"learning_rate": 4.937067337948396e-05,
|
| 85 |
+
"loss": 2.5532,
|
| 86 |
+
"step": 120
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.08,
|
| 90 |
+
"learning_rate": 4.9056010069225936e-05,
|
| 91 |
+
"loss": 2.5329,
|
| 92 |
+
"step": 130
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.08,
|
| 96 |
+
"learning_rate": 4.874134675896791e-05,
|
| 97 |
+
"loss": 2.4838,
|
| 98 |
+
"step": 140
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.09,
|
| 102 |
+
"learning_rate": 4.842668344870988e-05,
|
| 103 |
+
"loss": 2.5586,
|
| 104 |
+
"step": 150
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.09,
|
| 108 |
+
"learning_rate": 4.8112020138451856e-05,
|
| 109 |
+
"loss": 2.5078,
|
| 110 |
+
"step": 160
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.1,
|
| 114 |
+
"learning_rate": 4.7797356828193834e-05,
|
| 115 |
+
"loss": 2.5224,
|
| 116 |
+
"step": 170
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.11,
|
| 120 |
+
"learning_rate": 4.748269351793581e-05,
|
| 121 |
+
"loss": 2.5317,
|
| 122 |
+
"step": 180
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.11,
|
| 126 |
+
"learning_rate": 4.716803020767779e-05,
|
| 127 |
+
"loss": 2.5802,
|
| 128 |
+
"step": 190
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.12,
|
| 132 |
+
"learning_rate": 4.685336689741977e-05,
|
| 133 |
+
"loss": 2.4744,
|
| 134 |
+
"step": 200
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.12,
|
| 138 |
+
"learning_rate": 4.653870358716174e-05,
|
| 139 |
+
"loss": 2.549,
|
| 140 |
+
"step": 210
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.13,
|
| 144 |
+
"learning_rate": 4.622404027690371e-05,
|
| 145 |
+
"loss": 2.5513,
|
| 146 |
+
"step": 220
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.14,
|
| 150 |
+
"learning_rate": 4.590937696664569e-05,
|
| 151 |
+
"loss": 2.4999,
|
| 152 |
+
"step": 230
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.14,
|
| 156 |
+
"learning_rate": 4.5594713656387665e-05,
|
| 157 |
+
"loss": 2.5966,
|
| 158 |
+
"step": 240
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.15,
|
| 162 |
+
"learning_rate": 4.528005034612964e-05,
|
| 163 |
+
"loss": 2.5307,
|
| 164 |
+
"step": 250
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.15,
|
| 168 |
+
"learning_rate": 4.496538703587162e-05,
|
| 169 |
+
"loss": 2.5179,
|
| 170 |
+
"step": 260
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.16,
|
| 174 |
+
"learning_rate": 4.46507237256136e-05,
|
| 175 |
+
"loss": 2.5107,
|
| 176 |
+
"step": 270
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.17,
|
| 180 |
+
"learning_rate": 4.433606041535558e-05,
|
| 181 |
+
"loss": 2.473,
|
| 182 |
+
"step": 280
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.17,
|
| 186 |
+
"learning_rate": 4.402139710509755e-05,
|
| 187 |
+
"loss": 2.5107,
|
| 188 |
+
"step": 290
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.18,
|
| 192 |
+
"learning_rate": 4.370673379483952e-05,
|
| 193 |
+
"loss": 2.5232,
|
| 194 |
+
"step": 300
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.18,
|
| 198 |
+
"learning_rate": 4.33920704845815e-05,
|
| 199 |
+
"loss": 2.5239,
|
| 200 |
+
"step": 310
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.19,
|
| 204 |
+
"learning_rate": 4.3077407174323475e-05,
|
| 205 |
+
"loss": 2.559,
|
| 206 |
+
"step": 320
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.2,
|
| 210 |
+
"learning_rate": 4.276274386406545e-05,
|
| 211 |
+
"loss": 2.5414,
|
| 212 |
+
"step": 330
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.2,
|
| 216 |
+
"learning_rate": 4.244808055380743e-05,
|
| 217 |
+
"loss": 2.5183,
|
| 218 |
+
"step": 340
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.21,
|
| 222 |
+
"learning_rate": 4.213341724354941e-05,
|
| 223 |
+
"loss": 2.5468,
|
| 224 |
+
"step": 350
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.21,
|
| 228 |
+
"learning_rate": 4.181875393329138e-05,
|
| 229 |
+
"loss": 2.5972,
|
| 230 |
+
"step": 360
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.22,
|
| 234 |
+
"learning_rate": 4.150409062303336e-05,
|
| 235 |
+
"loss": 2.5869,
|
| 236 |
+
"step": 370
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.22,
|
| 240 |
+
"learning_rate": 4.118942731277533e-05,
|
| 241 |
+
"loss": 2.5745,
|
| 242 |
+
"step": 380
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.23,
|
| 246 |
+
"learning_rate": 4.0874764002517306e-05,
|
| 247 |
+
"loss": 2.6045,
|
| 248 |
+
"step": 390
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.24,
|
| 252 |
+
"learning_rate": 4.0560100692259284e-05,
|
| 253 |
+
"loss": 2.5491,
|
| 254 |
+
"step": 400
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.24,
|
| 258 |
+
"learning_rate": 4.024543738200126e-05,
|
| 259 |
+
"loss": 2.5674,
|
| 260 |
+
"step": 410
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.25,
|
| 264 |
+
"learning_rate": 3.993077407174324e-05,
|
| 265 |
+
"loss": 2.5299,
|
| 266 |
+
"step": 420
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.25,
|
| 270 |
+
"learning_rate": 3.961611076148522e-05,
|
| 271 |
+
"loss": 2.6367,
|
| 272 |
+
"step": 430
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.26,
|
| 276 |
+
"learning_rate": 3.930144745122719e-05,
|
| 277 |
+
"loss": 2.5684,
|
| 278 |
+
"step": 440
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.27,
|
| 282 |
+
"learning_rate": 3.898678414096916e-05,
|
| 283 |
+
"loss": 2.5414,
|
| 284 |
+
"step": 450
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.27,
|
| 288 |
+
"learning_rate": 3.867212083071114e-05,
|
| 289 |
+
"loss": 2.5335,
|
| 290 |
+
"step": 460
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.28,
|
| 294 |
+
"learning_rate": 3.8357457520453115e-05,
|
| 295 |
+
"loss": 2.5569,
|
| 296 |
+
"step": 470
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.28,
|
| 300 |
+
"learning_rate": 3.804279421019509e-05,
|
| 301 |
+
"loss": 2.602,
|
| 302 |
+
"step": 480
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.29,
|
| 306 |
+
"learning_rate": 3.772813089993707e-05,
|
| 307 |
+
"loss": 2.5796,
|
| 308 |
+
"step": 490
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.3,
|
| 312 |
+
"learning_rate": 3.741346758967905e-05,
|
| 313 |
+
"loss": 2.6123,
|
| 314 |
+
"step": 500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.3,
|
| 318 |
+
"learning_rate": 3.7098804279421027e-05,
|
| 319 |
+
"loss": 2.5303,
|
| 320 |
+
"step": 510
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.31,
|
| 324 |
+
"learning_rate": 3.6784140969163e-05,
|
| 325 |
+
"loss": 2.6367,
|
| 326 |
+
"step": 520
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.31,
|
| 330 |
+
"learning_rate": 3.646947765890497e-05,
|
| 331 |
+
"loss": 2.5969,
|
| 332 |
+
"step": 530
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.32,
|
| 336 |
+
"learning_rate": 3.615481434864695e-05,
|
| 337 |
+
"loss": 2.6128,
|
| 338 |
+
"step": 540
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.33,
|
| 342 |
+
"learning_rate": 3.5840151038388924e-05,
|
| 343 |
+
"loss": 2.6029,
|
| 344 |
+
"step": 550
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.33,
|
| 348 |
+
"learning_rate": 3.55254877281309e-05,
|
| 349 |
+
"loss": 2.5175,
|
| 350 |
+
"step": 560
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.34,
|
| 354 |
+
"learning_rate": 3.521082441787288e-05,
|
| 355 |
+
"loss": 2.5812,
|
| 356 |
+
"step": 570
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.34,
|
| 360 |
+
"learning_rate": 3.489616110761486e-05,
|
| 361 |
+
"loss": 2.6197,
|
| 362 |
+
"step": 580
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.35,
|
| 366 |
+
"learning_rate": 3.458149779735683e-05,
|
| 367 |
+
"loss": 2.6282,
|
| 368 |
+
"step": 590
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.36,
|
| 372 |
+
"learning_rate": 3.42668344870988e-05,
|
| 373 |
+
"loss": 2.6018,
|
| 374 |
+
"step": 600
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.36,
|
| 378 |
+
"learning_rate": 3.395217117684078e-05,
|
| 379 |
+
"loss": 2.5908,
|
| 380 |
+
"step": 610
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.37,
|
| 384 |
+
"learning_rate": 3.3637507866582756e-05,
|
| 385 |
+
"loss": 2.6068,
|
| 386 |
+
"step": 620
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.37,
|
| 390 |
+
"learning_rate": 3.3322844556324734e-05,
|
| 391 |
+
"loss": 2.5934,
|
| 392 |
+
"step": 630
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.38,
|
| 396 |
+
"learning_rate": 3.300818124606671e-05,
|
| 397 |
+
"loss": 2.6117,
|
| 398 |
+
"step": 640
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.38,
|
| 402 |
+
"learning_rate": 3.269351793580869e-05,
|
| 403 |
+
"loss": 2.5347,
|
| 404 |
+
"step": 650
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.39,
|
| 408 |
+
"learning_rate": 3.237885462555067e-05,
|
| 409 |
+
"loss": 2.6547,
|
| 410 |
+
"step": 660
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.4,
|
| 414 |
+
"learning_rate": 3.206419131529264e-05,
|
| 415 |
+
"loss": 2.6311,
|
| 416 |
+
"step": 670
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.4,
|
| 420 |
+
"learning_rate": 3.174952800503461e-05,
|
| 421 |
+
"loss": 2.6168,
|
| 422 |
+
"step": 680
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.41,
|
| 426 |
+
"learning_rate": 3.143486469477659e-05,
|
| 427 |
+
"loss": 2.5944,
|
| 428 |
+
"step": 690
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.41,
|
| 432 |
+
"learning_rate": 3.1120201384518565e-05,
|
| 433 |
+
"loss": 2.5865,
|
| 434 |
+
"step": 700
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.42,
|
| 438 |
+
"learning_rate": 3.080553807426054e-05,
|
| 439 |
+
"loss": 2.6284,
|
| 440 |
+
"step": 710
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.43,
|
| 444 |
+
"learning_rate": 3.049087476400252e-05,
|
| 445 |
+
"loss": 2.5945,
|
| 446 |
+
"step": 720
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.43,
|
| 450 |
+
"learning_rate": 3.0176211453744495e-05,
|
| 451 |
+
"loss": 2.5627,
|
| 452 |
+
"step": 730
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.44,
|
| 456 |
+
"learning_rate": 2.9861548143486473e-05,
|
| 457 |
+
"loss": 2.5967,
|
| 458 |
+
"step": 740
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.44,
|
| 462 |
+
"learning_rate": 2.9546884833228444e-05,
|
| 463 |
+
"loss": 2.6523,
|
| 464 |
+
"step": 750
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.45,
|
| 468 |
+
"learning_rate": 2.9232221522970422e-05,
|
| 469 |
+
"loss": 2.6584,
|
| 470 |
+
"step": 760
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.46,
|
| 474 |
+
"learning_rate": 2.8917558212712397e-05,
|
| 475 |
+
"loss": 2.6243,
|
| 476 |
+
"step": 770
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.46,
|
| 480 |
+
"learning_rate": 2.8602894902454374e-05,
|
| 481 |
+
"loss": 2.5734,
|
| 482 |
+
"step": 780
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.47,
|
| 486 |
+
"learning_rate": 2.8288231592196352e-05,
|
| 487 |
+
"loss": 2.5459,
|
| 488 |
+
"step": 790
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.47,
|
| 492 |
+
"learning_rate": 2.7973568281938327e-05,
|
| 493 |
+
"loss": 2.5985,
|
| 494 |
+
"step": 800
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.48,
|
| 498 |
+
"learning_rate": 2.7658904971680305e-05,
|
| 499 |
+
"loss": 2.6552,
|
| 500 |
+
"step": 810
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.49,
|
| 504 |
+
"learning_rate": 2.7344241661422282e-05,
|
| 505 |
+
"loss": 2.5976,
|
| 506 |
+
"step": 820
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.49,
|
| 510 |
+
"learning_rate": 2.7029578351164254e-05,
|
| 511 |
+
"loss": 2.5838,
|
| 512 |
+
"step": 830
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.5,
|
| 516 |
+
"learning_rate": 2.671491504090623e-05,
|
| 517 |
+
"loss": 2.6057,
|
| 518 |
+
"step": 840
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 0.5,
|
| 522 |
+
"learning_rate": 2.6400251730648206e-05,
|
| 523 |
+
"loss": 2.6784,
|
| 524 |
+
"step": 850
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 0.51,
|
| 528 |
+
"learning_rate": 2.6085588420390184e-05,
|
| 529 |
+
"loss": 2.6413,
|
| 530 |
+
"step": 860
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.51,
|
| 534 |
+
"learning_rate": 2.577092511013216e-05,
|
| 535 |
+
"loss": 2.5734,
|
| 536 |
+
"step": 870
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.52,
|
| 540 |
+
"learning_rate": 2.5456261799874136e-05,
|
| 541 |
+
"loss": 2.6205,
|
| 542 |
+
"step": 880
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.53,
|
| 546 |
+
"learning_rate": 2.5141598489616114e-05,
|
| 547 |
+
"loss": 2.6198,
|
| 548 |
+
"step": 890
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.53,
|
| 552 |
+
"learning_rate": 2.482693517935809e-05,
|
| 553 |
+
"loss": 2.5844,
|
| 554 |
+
"step": 900
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.54,
|
| 558 |
+
"learning_rate": 2.4512271869100066e-05,
|
| 559 |
+
"loss": 2.6097,
|
| 560 |
+
"step": 910
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.54,
|
| 564 |
+
"learning_rate": 2.419760855884204e-05,
|
| 565 |
+
"loss": 2.6332,
|
| 566 |
+
"step": 920
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 0.55,
|
| 570 |
+
"learning_rate": 2.3882945248584015e-05,
|
| 571 |
+
"loss": 2.6285,
|
| 572 |
+
"step": 930
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.56,
|
| 576 |
+
"learning_rate": 2.3568281938325993e-05,
|
| 577 |
+
"loss": 2.668,
|
| 578 |
+
"step": 940
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.56,
|
| 582 |
+
"learning_rate": 2.3253618628067967e-05,
|
| 583 |
+
"loss": 2.6187,
|
| 584 |
+
"step": 950
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.57,
|
| 588 |
+
"learning_rate": 2.2938955317809942e-05,
|
| 589 |
+
"loss": 2.6292,
|
| 590 |
+
"step": 960
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.57,
|
| 594 |
+
"learning_rate": 2.262429200755192e-05,
|
| 595 |
+
"loss": 2.6736,
|
| 596 |
+
"step": 970
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.58,
|
| 600 |
+
"learning_rate": 2.2309628697293898e-05,
|
| 601 |
+
"loss": 2.6426,
|
| 602 |
+
"step": 980
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.59,
|
| 606 |
+
"learning_rate": 2.1994965387035872e-05,
|
| 607 |
+
"loss": 2.6719,
|
| 608 |
+
"step": 990
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.59,
|
| 612 |
+
"learning_rate": 2.1680302076777847e-05,
|
| 613 |
+
"loss": 2.6793,
|
| 614 |
+
"step": 1000
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.6,
|
| 618 |
+
"learning_rate": 2.1365638766519824e-05,
|
| 619 |
+
"loss": 2.7038,
|
| 620 |
+
"step": 1010
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.6,
|
| 624 |
+
"learning_rate": 2.1050975456261802e-05,
|
| 625 |
+
"loss": 2.6515,
|
| 626 |
+
"step": 1020
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.61,
|
| 630 |
+
"learning_rate": 2.0736312146003777e-05,
|
| 631 |
+
"loss": 2.6198,
|
| 632 |
+
"step": 1030
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.62,
|
| 636 |
+
"learning_rate": 2.042164883574575e-05,
|
| 637 |
+
"loss": 2.6505,
|
| 638 |
+
"step": 1040
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 0.62,
|
| 642 |
+
"learning_rate": 2.010698552548773e-05,
|
| 643 |
+
"loss": 2.6638,
|
| 644 |
+
"step": 1050
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 0.63,
|
| 648 |
+
"learning_rate": 1.9792322215229707e-05,
|
| 649 |
+
"loss": 2.6602,
|
| 650 |
+
"step": 1060
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 0.63,
|
| 654 |
+
"learning_rate": 1.947765890497168e-05,
|
| 655 |
+
"loss": 2.599,
|
| 656 |
+
"step": 1070
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.64,
|
| 660 |
+
"learning_rate": 1.9162995594713656e-05,
|
| 661 |
+
"loss": 2.6421,
|
| 662 |
+
"step": 1080
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.65,
|
| 666 |
+
"learning_rate": 1.8848332284455634e-05,
|
| 667 |
+
"loss": 2.6311,
|
| 668 |
+
"step": 1090
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.65,
|
| 672 |
+
"learning_rate": 1.853366897419761e-05,
|
| 673 |
+
"loss": 2.617,
|
| 674 |
+
"step": 1100
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.66,
|
| 678 |
+
"learning_rate": 1.8219005663939586e-05,
|
| 679 |
+
"loss": 2.7233,
|
| 680 |
+
"step": 1110
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.66,
|
| 684 |
+
"learning_rate": 1.790434235368156e-05,
|
| 685 |
+
"loss": 2.6468,
|
| 686 |
+
"step": 1120
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 0.67,
|
| 690 |
+
"learning_rate": 1.7589679043423538e-05,
|
| 691 |
+
"loss": 2.6206,
|
| 692 |
+
"step": 1130
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 0.67,
|
| 696 |
+
"learning_rate": 1.7275015733165513e-05,
|
| 697 |
+
"loss": 2.6542,
|
| 698 |
+
"step": 1140
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 0.68,
|
| 702 |
+
"learning_rate": 1.696035242290749e-05,
|
| 703 |
+
"loss": 2.6512,
|
| 704 |
+
"step": 1150
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.69,
|
| 708 |
+
"learning_rate": 1.6645689112649465e-05,
|
| 709 |
+
"loss": 2.6035,
|
| 710 |
+
"step": 1160
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 0.69,
|
| 714 |
+
"learning_rate": 1.6331025802391443e-05,
|
| 715 |
+
"loss": 2.6696,
|
| 716 |
+
"step": 1170
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 0.7,
|
| 720 |
+
"learning_rate": 1.6016362492133417e-05,
|
| 721 |
+
"loss": 2.6677,
|
| 722 |
+
"step": 1180
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 0.7,
|
| 726 |
+
"learning_rate": 1.5701699181875392e-05,
|
| 727 |
+
"loss": 2.6828,
|
| 728 |
+
"step": 1190
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 0.71,
|
| 732 |
+
"learning_rate": 1.538703587161737e-05,
|
| 733 |
+
"loss": 2.6724,
|
| 734 |
+
"step": 1200
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 0.72,
|
| 738 |
+
"learning_rate": 1.5072372561359346e-05,
|
| 739 |
+
"loss": 2.6375,
|
| 740 |
+
"step": 1210
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 0.72,
|
| 744 |
+
"learning_rate": 1.4757709251101324e-05,
|
| 745 |
+
"loss": 2.6768,
|
| 746 |
+
"step": 1220
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 0.73,
|
| 750 |
+
"learning_rate": 1.4443045940843297e-05,
|
| 751 |
+
"loss": 2.6233,
|
| 752 |
+
"step": 1230
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 0.73,
|
| 756 |
+
"learning_rate": 1.4128382630585274e-05,
|
| 757 |
+
"loss": 2.6259,
|
| 758 |
+
"step": 1240
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 0.74,
|
| 762 |
+
"learning_rate": 1.381371932032725e-05,
|
| 763 |
+
"loss": 2.6801,
|
| 764 |
+
"step": 1250
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 0.75,
|
| 768 |
+
"learning_rate": 1.3499056010069228e-05,
|
| 769 |
+
"loss": 2.7294,
|
| 770 |
+
"step": 1260
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 0.75,
|
| 774 |
+
"learning_rate": 1.3184392699811201e-05,
|
| 775 |
+
"loss": 2.7005,
|
| 776 |
+
"step": 1270
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 0.76,
|
| 780 |
+
"learning_rate": 1.2869729389553179e-05,
|
| 781 |
+
"loss": 2.7266,
|
| 782 |
+
"step": 1280
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 0.76,
|
| 786 |
+
"learning_rate": 1.2555066079295155e-05,
|
| 787 |
+
"loss": 2.6862,
|
| 788 |
+
"step": 1290
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 0.77,
|
| 792 |
+
"learning_rate": 1.2240402769037131e-05,
|
| 793 |
+
"loss": 2.6623,
|
| 794 |
+
"step": 1300
|
| 795 |
+
},
|
| 796 |
+
{
|
| 797 |
+
"epoch": 0.78,
|
| 798 |
+
"learning_rate": 1.1925739458779107e-05,
|
| 799 |
+
"loss": 2.731,
|
| 800 |
+
"step": 1310
|
| 801 |
+
},
|
| 802 |
+
{
|
| 803 |
+
"epoch": 0.78,
|
| 804 |
+
"learning_rate": 1.1611076148521084e-05,
|
| 805 |
+
"loss": 2.6724,
|
| 806 |
+
"step": 1320
|
| 807 |
+
},
|
| 808 |
+
{
|
| 809 |
+
"epoch": 0.79,
|
| 810 |
+
"learning_rate": 1.129641283826306e-05,
|
| 811 |
+
"loss": 2.64,
|
| 812 |
+
"step": 1330
|
| 813 |
+
},
|
| 814 |
+
{
|
| 815 |
+
"epoch": 0.79,
|
| 816 |
+
"learning_rate": 1.0981749528005034e-05,
|
| 817 |
+
"loss": 2.7042,
|
| 818 |
+
"step": 1340
|
| 819 |
+
},
|
| 820 |
+
{
|
| 821 |
+
"epoch": 0.8,
|
| 822 |
+
"learning_rate": 1.0667086217747012e-05,
|
| 823 |
+
"loss": 2.6966,
|
| 824 |
+
"step": 1350
|
| 825 |
+
},
|
| 826 |
+
{
|
| 827 |
+
"epoch": 0.8,
|
| 828 |
+
"learning_rate": 1.0352422907488987e-05,
|
| 829 |
+
"loss": 2.7149,
|
| 830 |
+
"step": 1360
|
| 831 |
+
},
|
| 832 |
+
{
|
| 833 |
+
"epoch": 0.81,
|
| 834 |
+
"learning_rate": 1.0037759597230964e-05,
|
| 835 |
+
"loss": 2.7139,
|
| 836 |
+
"step": 1370
|
| 837 |
+
},
|
| 838 |
+
{
|
| 839 |
+
"epoch": 0.82,
|
| 840 |
+
"learning_rate": 9.723096286972939e-06,
|
| 841 |
+
"loss": 2.6948,
|
| 842 |
+
"step": 1380
|
| 843 |
+
},
|
| 844 |
+
{
|
| 845 |
+
"epoch": 0.82,
|
| 846 |
+
"learning_rate": 9.408432976714915e-06,
|
| 847 |
+
"loss": 2.7423,
|
| 848 |
+
"step": 1390
|
| 849 |
+
},
|
| 850 |
+
{
|
| 851 |
+
"epoch": 0.83,
|
| 852 |
+
"learning_rate": 9.093769666456891e-06,
|
| 853 |
+
"loss": 2.6981,
|
| 854 |
+
"step": 1400
|
| 855 |
+
},
|
| 856 |
+
{
|
| 857 |
+
"epoch": 0.83,
|
| 858 |
+
"learning_rate": 8.779106356198867e-06,
|
| 859 |
+
"loss": 2.7368,
|
| 860 |
+
"step": 1410
|
| 861 |
+
},
|
| 862 |
+
{
|
| 863 |
+
"epoch": 0.84,
|
| 864 |
+
"learning_rate": 8.464443045940844e-06,
|
| 865 |
+
"loss": 2.6836,
|
| 866 |
+
"step": 1420
|
| 867 |
+
},
|
| 868 |
+
{
|
| 869 |
+
"epoch": 0.85,
|
| 870 |
+
"learning_rate": 8.14977973568282e-06,
|
| 871 |
+
"loss": 2.7165,
|
| 872 |
+
"step": 1430
|
| 873 |
+
},
|
| 874 |
+
{
|
| 875 |
+
"epoch": 0.85,
|
| 876 |
+
"learning_rate": 7.835116425424796e-06,
|
| 877 |
+
"loss": 2.7344,
|
| 878 |
+
"step": 1440
|
| 879 |
+
},
|
| 880 |
+
{
|
| 881 |
+
"epoch": 0.86,
|
| 882 |
+
"learning_rate": 7.520453115166771e-06,
|
| 883 |
+
"loss": 2.7199,
|
| 884 |
+
"step": 1450
|
| 885 |
+
},
|
| 886 |
+
{
|
| 887 |
+
"epoch": 0.86,
|
| 888 |
+
"learning_rate": 7.205789804908748e-06,
|
| 889 |
+
"loss": 2.6674,
|
| 890 |
+
"step": 1460
|
| 891 |
+
},
|
| 892 |
+
{
|
| 893 |
+
"epoch": 0.87,
|
| 894 |
+
"learning_rate": 6.8911264946507235e-06,
|
| 895 |
+
"loss": 2.755,
|
| 896 |
+
"step": 1470
|
| 897 |
+
},
|
| 898 |
+
{
|
| 899 |
+
"epoch": 0.88,
|
| 900 |
+
"learning_rate": 6.5764631843927005e-06,
|
| 901 |
+
"loss": 2.6625,
|
| 902 |
+
"step": 1480
|
| 903 |
+
},
|
| 904 |
+
{
|
| 905 |
+
"epoch": 0.88,
|
| 906 |
+
"learning_rate": 6.261799874134676e-06,
|
| 907 |
+
"loss": 2.712,
|
| 908 |
+
"step": 1490
|
| 909 |
+
},
|
| 910 |
+
{
|
| 911 |
+
"epoch": 0.89,
|
| 912 |
+
"learning_rate": 5.947136563876652e-06,
|
| 913 |
+
"loss": 2.7506,
|
| 914 |
+
"step": 1500
|
| 915 |
+
},
|
| 916 |
+
{
|
| 917 |
+
"epoch": 0.89,
|
| 918 |
+
"learning_rate": 5.632473253618628e-06,
|
| 919 |
+
"loss": 2.7283,
|
| 920 |
+
"step": 1510
|
| 921 |
+
},
|
| 922 |
+
{
|
| 923 |
+
"epoch": 0.9,
|
| 924 |
+
"learning_rate": 5.317809943360604e-06,
|
| 925 |
+
"loss": 2.7416,
|
| 926 |
+
"step": 1520
|
| 927 |
+
},
|
| 928 |
+
{
|
| 929 |
+
"epoch": 0.91,
|
| 930 |
+
"learning_rate": 5.0031466331025804e-06,
|
| 931 |
+
"loss": 2.7262,
|
| 932 |
+
"step": 1530
|
| 933 |
+
},
|
| 934 |
+
{
|
| 935 |
+
"epoch": 0.91,
|
| 936 |
+
"learning_rate": 4.688483322844557e-06,
|
| 937 |
+
"loss": 2.7172,
|
| 938 |
+
"step": 1540
|
| 939 |
+
},
|
| 940 |
+
{
|
| 941 |
+
"epoch": 0.92,
|
| 942 |
+
"learning_rate": 4.373820012586533e-06,
|
| 943 |
+
"loss": 2.7281,
|
| 944 |
+
"step": 1550
|
| 945 |
+
},
|
| 946 |
+
{
|
| 947 |
+
"epoch": 0.92,
|
| 948 |
+
"learning_rate": 4.059156702328509e-06,
|
| 949 |
+
"loss": 2.7196,
|
| 950 |
+
"step": 1560
|
| 951 |
+
},
|
| 952 |
+
{
|
| 953 |
+
"epoch": 0.93,
|
| 954 |
+
"learning_rate": 3.7444933920704846e-06,
|
| 955 |
+
"loss": 2.7212,
|
| 956 |
+
"step": 1570
|
| 957 |
+
},
|
| 958 |
+
{
|
| 959 |
+
"epoch": 0.94,
|
| 960 |
+
"learning_rate": 3.429830081812461e-06,
|
| 961 |
+
"loss": 2.6828,
|
| 962 |
+
"step": 1580
|
| 963 |
+
},
|
| 964 |
+
{
|
| 965 |
+
"epoch": 0.94,
|
| 966 |
+
"learning_rate": 3.115166771554437e-06,
|
| 967 |
+
"loss": 2.6544,
|
| 968 |
+
"step": 1590
|
| 969 |
+
},
|
| 970 |
+
{
|
| 971 |
+
"epoch": 0.95,
|
| 972 |
+
"learning_rate": 2.800503461296413e-06,
|
| 973 |
+
"loss": 2.7682,
|
| 974 |
+
"step": 1600
|
| 975 |
+
},
|
| 976 |
+
{
|
| 977 |
+
"epoch": 0.95,
|
| 978 |
+
"learning_rate": 2.4858401510383893e-06,
|
| 979 |
+
"loss": 2.7133,
|
| 980 |
+
"step": 1610
|
| 981 |
+
},
|
| 982 |
+
{
|
| 983 |
+
"epoch": 0.96,
|
| 984 |
+
"learning_rate": 2.1711768407803654e-06,
|
| 985 |
+
"loss": 2.7196,
|
| 986 |
+
"step": 1620
|
| 987 |
+
},
|
| 988 |
+
{
|
| 989 |
+
"epoch": 0.96,
|
| 990 |
+
"learning_rate": 1.8565135305223414e-06,
|
| 991 |
+
"loss": 2.7418,
|
| 992 |
+
"step": 1630
|
| 993 |
+
},
|
| 994 |
+
{
|
| 995 |
+
"epoch": 0.97,
|
| 996 |
+
"learning_rate": 1.5418502202643173e-06,
|
| 997 |
+
"loss": 2.7931,
|
| 998 |
+
"step": 1640
|
| 999 |
+
},
|
| 1000 |
+
{
|
| 1001 |
+
"epoch": 0.98,
|
| 1002 |
+
"learning_rate": 1.2271869100062933e-06,
|
| 1003 |
+
"loss": 2.7782,
|
| 1004 |
+
"step": 1650
|
| 1005 |
+
},
|
| 1006 |
+
{
|
| 1007 |
+
"epoch": 0.98,
|
| 1008 |
+
"learning_rate": 9.125235997482694e-07,
|
| 1009 |
+
"loss": 2.7426,
|
| 1010 |
+
"step": 1660
|
| 1011 |
+
},
|
| 1012 |
+
{
|
| 1013 |
+
"epoch": 0.99,
|
| 1014 |
+
"learning_rate": 5.978602894902455e-07,
|
| 1015 |
+
"loss": 2.7689,
|
| 1016 |
+
"step": 1670
|
| 1017 |
+
},
|
| 1018 |
+
{
|
| 1019 |
+
"epoch": 0.99,
|
| 1020 |
+
"learning_rate": 2.8319697923222154e-07,
|
| 1021 |
+
"loss": 2.7109,
|
| 1022 |
+
"step": 1680
|
| 1023 |
+
},
|
| 1024 |
+
{
|
| 1025 |
+
"epoch": 1.0,
|
| 1026 |
+
"step": 1689,
|
| 1027 |
+
"total_flos": 5.644869713893786e+16,
|
| 1028 |
+
"train_loss": 2.6318450465295498,
|
| 1029 |
+
"train_runtime": 20073.5525,
|
| 1030 |
+
"train_samples_per_second": 0.337,
|
| 1031 |
+
"train_steps_per_second": 0.084
|
| 1032 |
+
}
|
| 1033 |
+
],
|
| 1034 |
+
"max_steps": 1689,
|
| 1035 |
+
"num_train_epochs": 1,
|
| 1036 |
+
"total_flos": 5.644869713893786e+16,
|
| 1037 |
+
"trial_name": null,
|
| 1038 |
+
"trial_params": null
|
| 1039 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ec97157145336b121b780afc9d3598e4cbd83d4bbe6f044da0e708869321e4b7
|
| 3 |
+
size 4340
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|