Initial commit
Browse files- added_tokens.json +4 -0
- config.json +41 -0
- generation_config.json +6 -0
- merges.txt +0 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +6 -0
- tokenizer.json +0 -0
- tokenizer_config.json +10 -0
- trainer_state.json +802 -0
- vocab.json +0 -0
added_tokens.json
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"<|pad|>": 50257,
|
| 3 |
+
"<|unk|>": 50258
|
| 4 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,41 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "gpt2-xl",
|
| 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 |
+
"initializer_range": 0.02,
|
| 12 |
+
"layer_norm_epsilon": 1e-05,
|
| 13 |
+
"model_type": "gpt2",
|
| 14 |
+
"n_ctx": 1024,
|
| 15 |
+
"n_embd": 1600,
|
| 16 |
+
"n_head": 25,
|
| 17 |
+
"n_inner": null,
|
| 18 |
+
"n_layer": 48,
|
| 19 |
+
"n_positions": 1024,
|
| 20 |
+
"output_past": true,
|
| 21 |
+
"pad_token_id": 50256,
|
| 22 |
+
"reorder_and_upcast_attn": false,
|
| 23 |
+
"resid_pdrop": 0.1,
|
| 24 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 25 |
+
"scale_attn_weights": true,
|
| 26 |
+
"summary_activation": null,
|
| 27 |
+
"summary_first_dropout": 0.1,
|
| 28 |
+
"summary_proj_to_labels": true,
|
| 29 |
+
"summary_type": "cls_index",
|
| 30 |
+
"summary_use_proj": true,
|
| 31 |
+
"task_specific_params": {
|
| 32 |
+
"text-generation": {
|
| 33 |
+
"do_sample": true,
|
| 34 |
+
"max_length": 50
|
| 35 |
+
}
|
| 36 |
+
},
|
| 37 |
+
"torch_dtype": "bfloat16",
|
| 38 |
+
"transformers_version": "4.30.2",
|
| 39 |
+
"use_cache": false,
|
| 40 |
+
"vocab_size": 50257
|
| 41 |
+
}
|
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": "4.30.2"
|
| 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:c8f25965ed849b52021c733f0cc334c03651415612c2827a7a58f2b4d92729a7
|
| 3 |
+
size 6230631363
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<|endoftext|>",
|
| 3 |
+
"eos_token": "<|endoftext|>",
|
| 4 |
+
"pad_token": "<|endoftext|>",
|
| 5 |
+
"unk_token": "<|endoftext|>"
|
| 6 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"bos_token": "<|endoftext|>",
|
| 4 |
+
"clean_up_tokenization_spaces": true,
|
| 5 |
+
"eos_token": "<|endoftext|>",
|
| 6 |
+
"model_max_length": 1024,
|
| 7 |
+
"padding_side": "right",
|
| 8 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 9 |
+
"unk_token": "<|endoftext|>"
|
| 10 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,802 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 16.0,
|
| 5 |
+
"global_step": 16800,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.12,
|
| 12 |
+
"learning_rate": 5e-05,
|
| 13 |
+
"loss": 1.7753,
|
| 14 |
+
"step": 128
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.24,
|
| 18 |
+
"learning_rate": 5e-05,
|
| 19 |
+
"loss": 1.4026,
|
| 20 |
+
"step": 256
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.37,
|
| 24 |
+
"learning_rate": 5e-05,
|
| 25 |
+
"loss": 1.3331,
|
| 26 |
+
"step": 384
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.49,
|
| 30 |
+
"learning_rate": 5e-05,
|
| 31 |
+
"loss": 1.3015,
|
| 32 |
+
"step": 512
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.61,
|
| 36 |
+
"learning_rate": 5e-05,
|
| 37 |
+
"loss": 1.2594,
|
| 38 |
+
"step": 640
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.73,
|
| 42 |
+
"learning_rate": 5e-05,
|
| 43 |
+
"loss": 1.2452,
|
| 44 |
+
"step": 768
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.85,
|
| 48 |
+
"learning_rate": 5e-05,
|
| 49 |
+
"loss": 1.2219,
|
| 50 |
+
"step": 896
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.98,
|
| 54 |
+
"learning_rate": 5e-05,
|
| 55 |
+
"loss": 1.2057,
|
| 56 |
+
"step": 1024
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 1.1,
|
| 60 |
+
"learning_rate": 5e-05,
|
| 61 |
+
"loss": 1.086,
|
| 62 |
+
"step": 1152
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 1.22,
|
| 66 |
+
"learning_rate": 5e-05,
|
| 67 |
+
"loss": 1.046,
|
| 68 |
+
"step": 1280
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 1.34,
|
| 72 |
+
"learning_rate": 5e-05,
|
| 73 |
+
"loss": 1.0642,
|
| 74 |
+
"step": 1408
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.46,
|
| 78 |
+
"learning_rate": 5e-05,
|
| 79 |
+
"loss": 1.0602,
|
| 80 |
+
"step": 1536
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.58,
|
| 84 |
+
"learning_rate": 5e-05,
|
| 85 |
+
"loss": 1.0552,
|
| 86 |
+
"step": 1664
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 1.71,
|
| 90 |
+
"learning_rate": 5e-05,
|
| 91 |
+
"loss": 1.0565,
|
| 92 |
+
"step": 1792
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 1.83,
|
| 96 |
+
"learning_rate": 5e-05,
|
| 97 |
+
"loss": 1.0602,
|
| 98 |
+
"step": 1920
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.95,
|
| 102 |
+
"learning_rate": 5e-05,
|
| 103 |
+
"loss": 1.0534,
|
| 104 |
+
"step": 2048
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 2.07,
|
| 108 |
+
"learning_rate": 5e-05,
|
| 109 |
+
"loss": 0.9373,
|
| 110 |
+
"step": 2176
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 2.19,
|
| 114 |
+
"learning_rate": 5e-05,
|
| 115 |
+
"loss": 0.8638,
|
| 116 |
+
"step": 2304
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 2.32,
|
| 120 |
+
"learning_rate": 5e-05,
|
| 121 |
+
"loss": 0.8759,
|
| 122 |
+
"step": 2432
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 2.44,
|
| 126 |
+
"learning_rate": 5e-05,
|
| 127 |
+
"loss": 0.8775,
|
| 128 |
+
"step": 2560
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 2.56,
|
| 132 |
+
"learning_rate": 5e-05,
|
| 133 |
+
"loss": 0.8906,
|
| 134 |
+
"step": 2688
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 2.68,
|
| 138 |
+
"learning_rate": 5e-05,
|
| 139 |
+
"loss": 0.8947,
|
| 140 |
+
"step": 2816
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 2.8,
|
| 144 |
+
"learning_rate": 5e-05,
|
| 145 |
+
"loss": 0.8846,
|
| 146 |
+
"step": 2944
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 2.93,
|
| 150 |
+
"learning_rate": 5e-05,
|
| 151 |
+
"loss": 0.8891,
|
| 152 |
+
"step": 3072
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 3.05,
|
| 156 |
+
"learning_rate": 5e-05,
|
| 157 |
+
"loss": 0.8039,
|
| 158 |
+
"step": 3200
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 3.17,
|
| 162 |
+
"learning_rate": 5e-05,
|
| 163 |
+
"loss": 0.6736,
|
| 164 |
+
"step": 3328
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 3.29,
|
| 168 |
+
"learning_rate": 5e-05,
|
| 169 |
+
"loss": 0.6814,
|
| 170 |
+
"step": 3456
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 3.41,
|
| 174 |
+
"learning_rate": 5e-05,
|
| 175 |
+
"loss": 0.7016,
|
| 176 |
+
"step": 3584
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 3.54,
|
| 180 |
+
"learning_rate": 5e-05,
|
| 181 |
+
"loss": 0.7039,
|
| 182 |
+
"step": 3712
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 3.66,
|
| 186 |
+
"learning_rate": 5e-05,
|
| 187 |
+
"loss": 0.708,
|
| 188 |
+
"step": 3840
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 3.78,
|
| 192 |
+
"learning_rate": 5e-05,
|
| 193 |
+
"loss": 0.7175,
|
| 194 |
+
"step": 3968
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 3.9,
|
| 198 |
+
"learning_rate": 5e-05,
|
| 199 |
+
"loss": 0.7214,
|
| 200 |
+
"step": 4096
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 4.02,
|
| 204 |
+
"learning_rate": 5e-05,
|
| 205 |
+
"loss": 0.6764,
|
| 206 |
+
"step": 4224
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 4.14,
|
| 210 |
+
"learning_rate": 5e-05,
|
| 211 |
+
"loss": 0.4891,
|
| 212 |
+
"step": 4352
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 4.27,
|
| 216 |
+
"learning_rate": 5e-05,
|
| 217 |
+
"loss": 0.5006,
|
| 218 |
+
"step": 4480
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 4.39,
|
| 222 |
+
"learning_rate": 5e-05,
|
| 223 |
+
"loss": 0.5129,
|
| 224 |
+
"step": 4608
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 4.51,
|
| 228 |
+
"learning_rate": 5e-05,
|
| 229 |
+
"loss": 0.5221,
|
| 230 |
+
"step": 4736
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 4.63,
|
| 234 |
+
"learning_rate": 5e-05,
|
| 235 |
+
"loss": 0.5278,
|
| 236 |
+
"step": 4864
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 4.75,
|
| 240 |
+
"learning_rate": 5e-05,
|
| 241 |
+
"loss": 0.5413,
|
| 242 |
+
"step": 4992
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 4.88,
|
| 246 |
+
"learning_rate": 5e-05,
|
| 247 |
+
"loss": 0.5393,
|
| 248 |
+
"step": 5120
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 5.0,
|
| 252 |
+
"learning_rate": 5e-05,
|
| 253 |
+
"loss": 0.5398,
|
| 254 |
+
"step": 5248
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 5.12,
|
| 258 |
+
"learning_rate": 5e-05,
|
| 259 |
+
"loss": 0.3369,
|
| 260 |
+
"step": 5376
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 5.24,
|
| 264 |
+
"learning_rate": 5e-05,
|
| 265 |
+
"loss": 0.3417,
|
| 266 |
+
"step": 5504
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 5.36,
|
| 270 |
+
"learning_rate": 5e-05,
|
| 271 |
+
"loss": 0.3502,
|
| 272 |
+
"step": 5632
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 5.49,
|
| 276 |
+
"learning_rate": 5e-05,
|
| 277 |
+
"loss": 0.3593,
|
| 278 |
+
"step": 5760
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 5.61,
|
| 282 |
+
"learning_rate": 5e-05,
|
| 283 |
+
"loss": 0.3695,
|
| 284 |
+
"step": 5888
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 5.73,
|
| 288 |
+
"learning_rate": 5e-05,
|
| 289 |
+
"loss": 0.3764,
|
| 290 |
+
"step": 6016
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 5.85,
|
| 294 |
+
"learning_rate": 5e-05,
|
| 295 |
+
"loss": 0.3831,
|
| 296 |
+
"step": 6144
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 5.97,
|
| 300 |
+
"learning_rate": 5e-05,
|
| 301 |
+
"loss": 0.3891,
|
| 302 |
+
"step": 6272
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 6.1,
|
| 306 |
+
"learning_rate": 5e-05,
|
| 307 |
+
"loss": 0.2571,
|
| 308 |
+
"step": 6400
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 6.22,
|
| 312 |
+
"learning_rate": 5e-05,
|
| 313 |
+
"loss": 0.2232,
|
| 314 |
+
"step": 6528
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 6.34,
|
| 318 |
+
"learning_rate": 5e-05,
|
| 319 |
+
"loss": 0.2324,
|
| 320 |
+
"step": 6656
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 6.46,
|
| 324 |
+
"learning_rate": 5e-05,
|
| 325 |
+
"loss": 0.2421,
|
| 326 |
+
"step": 6784
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 6.58,
|
| 330 |
+
"learning_rate": 5e-05,
|
| 331 |
+
"loss": 0.2479,
|
| 332 |
+
"step": 6912
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 6.7,
|
| 336 |
+
"learning_rate": 5e-05,
|
| 337 |
+
"loss": 0.2584,
|
| 338 |
+
"step": 7040
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 6.83,
|
| 342 |
+
"learning_rate": 5e-05,
|
| 343 |
+
"loss": 0.2642,
|
| 344 |
+
"step": 7168
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 6.95,
|
| 348 |
+
"learning_rate": 5e-05,
|
| 349 |
+
"loss": 0.2706,
|
| 350 |
+
"step": 7296
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 7.07,
|
| 354 |
+
"learning_rate": 5e-05,
|
| 355 |
+
"loss": 0.203,
|
| 356 |
+
"step": 7424
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 7.19,
|
| 360 |
+
"learning_rate": 5e-05,
|
| 361 |
+
"loss": 0.1488,
|
| 362 |
+
"step": 7552
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 7.31,
|
| 366 |
+
"learning_rate": 5e-05,
|
| 367 |
+
"loss": 0.1654,
|
| 368 |
+
"step": 7680
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 7.44,
|
| 372 |
+
"learning_rate": 5e-05,
|
| 373 |
+
"loss": 0.1706,
|
| 374 |
+
"step": 7808
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 7.56,
|
| 378 |
+
"learning_rate": 5e-05,
|
| 379 |
+
"loss": 0.1799,
|
| 380 |
+
"step": 7936
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 7.68,
|
| 384 |
+
"learning_rate": 5e-05,
|
| 385 |
+
"loss": 0.1823,
|
| 386 |
+
"step": 8064
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 7.8,
|
| 390 |
+
"learning_rate": 5e-05,
|
| 391 |
+
"loss": 0.1867,
|
| 392 |
+
"step": 8192
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 7.92,
|
| 396 |
+
"learning_rate": 5e-05,
|
| 397 |
+
"loss": 0.1931,
|
| 398 |
+
"step": 8320
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 8.05,
|
| 402 |
+
"learning_rate": 5e-05,
|
| 403 |
+
"loss": 0.1635,
|
| 404 |
+
"step": 8448
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 8.17,
|
| 408 |
+
"learning_rate": 5e-05,
|
| 409 |
+
"loss": 0.1111,
|
| 410 |
+
"step": 8576
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 8.29,
|
| 414 |
+
"learning_rate": 5e-05,
|
| 415 |
+
"loss": 0.1158,
|
| 416 |
+
"step": 8704
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 8.41,
|
| 420 |
+
"learning_rate": 5e-05,
|
| 421 |
+
"loss": 0.1188,
|
| 422 |
+
"step": 8832
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 8.53,
|
| 426 |
+
"learning_rate": 5e-05,
|
| 427 |
+
"loss": 0.124,
|
| 428 |
+
"step": 8960
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 8.66,
|
| 432 |
+
"learning_rate": 5e-05,
|
| 433 |
+
"loss": 0.1295,
|
| 434 |
+
"step": 9088
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 8.78,
|
| 438 |
+
"learning_rate": 5e-05,
|
| 439 |
+
"loss": 0.1348,
|
| 440 |
+
"step": 9216
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 8.9,
|
| 444 |
+
"learning_rate": 5e-05,
|
| 445 |
+
"loss": 0.1386,
|
| 446 |
+
"step": 9344
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 9.02,
|
| 450 |
+
"learning_rate": 5e-05,
|
| 451 |
+
"loss": 0.1343,
|
| 452 |
+
"step": 9472
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 9.14,
|
| 456 |
+
"learning_rate": 5e-05,
|
| 457 |
+
"loss": 0.0847,
|
| 458 |
+
"step": 9600
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 9.26,
|
| 462 |
+
"learning_rate": 5e-05,
|
| 463 |
+
"loss": 0.0873,
|
| 464 |
+
"step": 9728
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 9.39,
|
| 468 |
+
"learning_rate": 5e-05,
|
| 469 |
+
"loss": 0.0915,
|
| 470 |
+
"step": 9856
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 9.51,
|
| 474 |
+
"learning_rate": 5e-05,
|
| 475 |
+
"loss": 0.0961,
|
| 476 |
+
"step": 9984
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 9.63,
|
| 480 |
+
"learning_rate": 5e-05,
|
| 481 |
+
"loss": 0.1003,
|
| 482 |
+
"step": 10112
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 9.75,
|
| 486 |
+
"learning_rate": 5e-05,
|
| 487 |
+
"loss": 0.1043,
|
| 488 |
+
"step": 10240
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 9.87,
|
| 492 |
+
"learning_rate": 5e-05,
|
| 493 |
+
"loss": 0.1094,
|
| 494 |
+
"step": 10368
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 10.0,
|
| 498 |
+
"learning_rate": 5e-05,
|
| 499 |
+
"loss": 0.1135,
|
| 500 |
+
"step": 10496
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 10.12,
|
| 504 |
+
"learning_rate": 5e-05,
|
| 505 |
+
"loss": 0.0723,
|
| 506 |
+
"step": 10624
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 10.24,
|
| 510 |
+
"learning_rate": 5e-05,
|
| 511 |
+
"loss": 0.0724,
|
| 512 |
+
"step": 10752
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 10.36,
|
| 516 |
+
"learning_rate": 5e-05,
|
| 517 |
+
"loss": 0.0759,
|
| 518 |
+
"step": 10880
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 10.48,
|
| 522 |
+
"learning_rate": 5e-05,
|
| 523 |
+
"loss": 0.0779,
|
| 524 |
+
"step": 11008
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 10.61,
|
| 528 |
+
"learning_rate": 5e-05,
|
| 529 |
+
"loss": 0.0817,
|
| 530 |
+
"step": 11136
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 10.73,
|
| 534 |
+
"learning_rate": 5e-05,
|
| 535 |
+
"loss": 0.0867,
|
| 536 |
+
"step": 11264
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 10.85,
|
| 540 |
+
"learning_rate": 5e-05,
|
| 541 |
+
"loss": 0.0908,
|
| 542 |
+
"step": 11392
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 10.97,
|
| 546 |
+
"learning_rate": 5e-05,
|
| 547 |
+
"loss": 0.0954,
|
| 548 |
+
"step": 11520
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 11.09,
|
| 552 |
+
"learning_rate": 5e-05,
|
| 553 |
+
"loss": 0.0691,
|
| 554 |
+
"step": 11648
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 11.22,
|
| 558 |
+
"learning_rate": 5e-05,
|
| 559 |
+
"loss": 0.0627,
|
| 560 |
+
"step": 11776
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 11.34,
|
| 564 |
+
"learning_rate": 5e-05,
|
| 565 |
+
"loss": 0.0649,
|
| 566 |
+
"step": 11904
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 11.46,
|
| 570 |
+
"learning_rate": 5e-05,
|
| 571 |
+
"loss": 0.0683,
|
| 572 |
+
"step": 12032
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 11.58,
|
| 576 |
+
"learning_rate": 5e-05,
|
| 577 |
+
"loss": 0.0721,
|
| 578 |
+
"step": 12160
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 11.7,
|
| 582 |
+
"learning_rate": 5e-05,
|
| 583 |
+
"loss": 0.0747,
|
| 584 |
+
"step": 12288
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 11.82,
|
| 588 |
+
"learning_rate": 5e-05,
|
| 589 |
+
"loss": 0.0768,
|
| 590 |
+
"step": 12416
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 11.95,
|
| 594 |
+
"learning_rate": 5e-05,
|
| 595 |
+
"loss": 0.0809,
|
| 596 |
+
"step": 12544
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 12.07,
|
| 600 |
+
"learning_rate": 5e-05,
|
| 601 |
+
"loss": 0.0681,
|
| 602 |
+
"step": 12672
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 12.19,
|
| 606 |
+
"learning_rate": 5e-05,
|
| 607 |
+
"loss": 0.0546,
|
| 608 |
+
"step": 12800
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 12.31,
|
| 612 |
+
"learning_rate": 5e-05,
|
| 613 |
+
"loss": 0.0561,
|
| 614 |
+
"step": 12928
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 12.43,
|
| 618 |
+
"learning_rate": 5e-05,
|
| 619 |
+
"loss": 0.059,
|
| 620 |
+
"step": 13056
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 12.56,
|
| 624 |
+
"learning_rate": 5e-05,
|
| 625 |
+
"loss": 0.0612,
|
| 626 |
+
"step": 13184
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 12.68,
|
| 630 |
+
"learning_rate": 5e-05,
|
| 631 |
+
"loss": 0.0644,
|
| 632 |
+
"step": 13312
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 12.8,
|
| 636 |
+
"learning_rate": 5e-05,
|
| 637 |
+
"loss": 0.0693,
|
| 638 |
+
"step": 13440
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 12.92,
|
| 642 |
+
"learning_rate": 5e-05,
|
| 643 |
+
"loss": 0.072,
|
| 644 |
+
"step": 13568
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 13.04,
|
| 648 |
+
"learning_rate": 5e-05,
|
| 649 |
+
"loss": 0.0663,
|
| 650 |
+
"step": 13696
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 13.17,
|
| 654 |
+
"learning_rate": 5e-05,
|
| 655 |
+
"loss": 0.0506,
|
| 656 |
+
"step": 13824
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 13.29,
|
| 660 |
+
"learning_rate": 5e-05,
|
| 661 |
+
"loss": 0.0512,
|
| 662 |
+
"step": 13952
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 13.41,
|
| 666 |
+
"learning_rate": 5e-05,
|
| 667 |
+
"loss": 0.0539,
|
| 668 |
+
"step": 14080
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 13.53,
|
| 672 |
+
"learning_rate": 5e-05,
|
| 673 |
+
"loss": 0.057,
|
| 674 |
+
"step": 14208
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 13.65,
|
| 678 |
+
"learning_rate": 5e-05,
|
| 679 |
+
"loss": 0.059,
|
| 680 |
+
"step": 14336
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 13.78,
|
| 684 |
+
"learning_rate": 5e-05,
|
| 685 |
+
"loss": 0.0614,
|
| 686 |
+
"step": 14464
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 13.9,
|
| 690 |
+
"learning_rate": 5e-05,
|
| 691 |
+
"loss": 0.0643,
|
| 692 |
+
"step": 14592
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 14.02,
|
| 696 |
+
"learning_rate": 5e-05,
|
| 697 |
+
"loss": 0.0645,
|
| 698 |
+
"step": 14720
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 14.14,
|
| 702 |
+
"learning_rate": 5e-05,
|
| 703 |
+
"loss": 0.0454,
|
| 704 |
+
"step": 14848
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 14.26,
|
| 708 |
+
"learning_rate": 5e-05,
|
| 709 |
+
"loss": 0.0465,
|
| 710 |
+
"step": 14976
|
| 711 |
+
},
|
| 712 |
+
{
|
| 713 |
+
"epoch": 14.38,
|
| 714 |
+
"learning_rate": 5e-05,
|
| 715 |
+
"loss": 0.0484,
|
| 716 |
+
"step": 15104
|
| 717 |
+
},
|
| 718 |
+
{
|
| 719 |
+
"epoch": 14.51,
|
| 720 |
+
"learning_rate": 5e-05,
|
| 721 |
+
"loss": 0.0508,
|
| 722 |
+
"step": 15232
|
| 723 |
+
},
|
| 724 |
+
{
|
| 725 |
+
"epoch": 14.63,
|
| 726 |
+
"learning_rate": 5e-05,
|
| 727 |
+
"loss": 0.053,
|
| 728 |
+
"step": 15360
|
| 729 |
+
},
|
| 730 |
+
{
|
| 731 |
+
"epoch": 14.75,
|
| 732 |
+
"learning_rate": 5e-05,
|
| 733 |
+
"loss": 0.0546,
|
| 734 |
+
"step": 15488
|
| 735 |
+
},
|
| 736 |
+
{
|
| 737 |
+
"epoch": 14.87,
|
| 738 |
+
"learning_rate": 5e-05,
|
| 739 |
+
"loss": 0.0566,
|
| 740 |
+
"step": 15616
|
| 741 |
+
},
|
| 742 |
+
{
|
| 743 |
+
"epoch": 14.99,
|
| 744 |
+
"learning_rate": 5e-05,
|
| 745 |
+
"loss": 0.0594,
|
| 746 |
+
"step": 15744
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 15.12,
|
| 750 |
+
"learning_rate": 5e-05,
|
| 751 |
+
"loss": 0.043,
|
| 752 |
+
"step": 15872
|
| 753 |
+
},
|
| 754 |
+
{
|
| 755 |
+
"epoch": 15.24,
|
| 756 |
+
"learning_rate": 5e-05,
|
| 757 |
+
"loss": 0.0433,
|
| 758 |
+
"step": 16000
|
| 759 |
+
},
|
| 760 |
+
{
|
| 761 |
+
"epoch": 15.36,
|
| 762 |
+
"learning_rate": 5e-05,
|
| 763 |
+
"loss": 0.0456,
|
| 764 |
+
"step": 16128
|
| 765 |
+
},
|
| 766 |
+
{
|
| 767 |
+
"epoch": 15.48,
|
| 768 |
+
"learning_rate": 5e-05,
|
| 769 |
+
"loss": 0.0474,
|
| 770 |
+
"step": 16256
|
| 771 |
+
},
|
| 772 |
+
{
|
| 773 |
+
"epoch": 15.6,
|
| 774 |
+
"learning_rate": 5e-05,
|
| 775 |
+
"loss": 0.049,
|
| 776 |
+
"step": 16384
|
| 777 |
+
},
|
| 778 |
+
{
|
| 779 |
+
"epoch": 15.73,
|
| 780 |
+
"learning_rate": 5e-05,
|
| 781 |
+
"loss": 0.0501,
|
| 782 |
+
"step": 16512
|
| 783 |
+
},
|
| 784 |
+
{
|
| 785 |
+
"epoch": 15.85,
|
| 786 |
+
"learning_rate": 5e-05,
|
| 787 |
+
"loss": 0.0522,
|
| 788 |
+
"step": 16640
|
| 789 |
+
},
|
| 790 |
+
{
|
| 791 |
+
"epoch": 15.97,
|
| 792 |
+
"learning_rate": 5e-05,
|
| 793 |
+
"loss": 0.0554,
|
| 794 |
+
"step": 16768
|
| 795 |
+
}
|
| 796 |
+
],
|
| 797 |
+
"max_steps": 21000,
|
| 798 |
+
"num_train_epochs": 20,
|
| 799 |
+
"total_flos": 3285446215892992.0,
|
| 800 |
+
"trial_name": null,
|
| 801 |
+
"trial_params": null
|
| 802 |
+
}
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|