commit from banalyst
Browse files- all_results.json +14 -0
- config.json +40 -0
- eval_results.json +9 -0
- merges.txt +0 -0
- pytorch_model.bin +3 -0
- special_tokens_map.json +1 -0
- tokenizer.json +0 -0
- tokenizer_config.json +1 -0
- train_results.json +8 -0
- trainer_state.json +667 -0
- training_args.bin +3 -0
- vocab.json +0 -0
all_results.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 2.9073383808135986,
|
| 4 |
+
"eval_runtime": 102.8484,
|
| 5 |
+
"eval_samples": 597,
|
| 6 |
+
"eval_samples_per_second": 5.805,
|
| 7 |
+
"eval_steps_per_second": 0.729,
|
| 8 |
+
"perplexity": 18.3080047238711,
|
| 9 |
+
"train_loss": 2.279561182284652,
|
| 10 |
+
"train_runtime": 34779.9602,
|
| 11 |
+
"train_samples": 5383,
|
| 12 |
+
"train_samples_per_second": 1.548,
|
| 13 |
+
"train_steps_per_second": 1.548
|
| 14 |
+
}
|
config.json
ADDED
|
@@ -0,0 +1,40 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "gpt2-medium",
|
| 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": 1024,
|
| 16 |
+
"n_embd": 1024,
|
| 17 |
+
"n_head": 16,
|
| 18 |
+
"n_inner": null,
|
| 19 |
+
"n_layer": 24,
|
| 20 |
+
"n_positions": 1024,
|
| 21 |
+
"n_special": 0,
|
| 22 |
+
"predict_special_tokens": true,
|
| 23 |
+
"resid_pdrop": 0.1,
|
| 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 |
+
"task_specific_params": {
|
| 31 |
+
"text-generation": {
|
| 32 |
+
"do_sample": true,
|
| 33 |
+
"max_length": 50
|
| 34 |
+
}
|
| 35 |
+
},
|
| 36 |
+
"torch_dtype": "float32",
|
| 37 |
+
"transformers_version": "4.9.0.dev0",
|
| 38 |
+
"use_cache": true,
|
| 39 |
+
"vocab_size": 50257
|
| 40 |
+
}
|
eval_results.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"eval_loss": 2.9073383808135986,
|
| 4 |
+
"eval_runtime": 102.8484,
|
| 5 |
+
"eval_samples": 597,
|
| 6 |
+
"eval_samples_per_second": 5.805,
|
| 7 |
+
"eval_steps_per_second": 0.729,
|
| 8 |
+
"perplexity": 18.3080047238711
|
| 9 |
+
}
|
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 sha256:2adc94001d8da0052485322726d1618f2720c21e67495deb863c354cd7ae90a1
|
| 3 |
+
size 1444581337
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "unk_token": "<|endoftext|>"}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{"unk_token": "<|endoftext|>", "bos_token": "<|endoftext|>", "eos_token": "<|endoftext|>", "add_prefix_space": false, "model_max_length": 1024, "special_tokens_map_file": null, "name_or_path": "gpt2-medium", "tokenizer_class": "GPT2Tokenizer"}
|
train_results.json
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"epoch": 10.0,
|
| 3 |
+
"train_loss": 2.279561182284652,
|
| 4 |
+
"train_runtime": 34779.9602,
|
| 5 |
+
"train_samples": 5383,
|
| 6 |
+
"train_samples_per_second": 1.548,
|
| 7 |
+
"train_steps_per_second": 1.548
|
| 8 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,667 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 10.0,
|
| 5 |
+
"global_step": 53830,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.09,
|
| 12 |
+
"learning_rate": 4.953836150845253e-05,
|
| 13 |
+
"loss": 3.2037,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.19,
|
| 18 |
+
"learning_rate": 4.9073936466654286e-05,
|
| 19 |
+
"loss": 3.166,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.28,
|
| 24 |
+
"learning_rate": 4.860951142485603e-05,
|
| 25 |
+
"loss": 3.1434,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.37,
|
| 30 |
+
"learning_rate": 4.814508638305778e-05,
|
| 31 |
+
"loss": 3.1104,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.46,
|
| 36 |
+
"learning_rate": 4.7680661341259524e-05,
|
| 37 |
+
"loss": 3.0987,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.56,
|
| 42 |
+
"learning_rate": 4.721623629946127e-05,
|
| 43 |
+
"loss": 3.0677,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.65,
|
| 48 |
+
"learning_rate": 4.6751811257663017e-05,
|
| 49 |
+
"loss": 3.0787,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.74,
|
| 54 |
+
"learning_rate": 4.628738621586476e-05,
|
| 55 |
+
"loss": 3.047,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.84,
|
| 60 |
+
"learning_rate": 4.582389002415011e-05,
|
| 61 |
+
"loss": 3.0487,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.93,
|
| 66 |
+
"learning_rate": 4.5359464982351854e-05,
|
| 67 |
+
"loss": 3.0396,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 1.02,
|
| 72 |
+
"learning_rate": 4.489596879063719e-05,
|
| 73 |
+
"loss": 2.983,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 1.11,
|
| 78 |
+
"learning_rate": 4.443154374883894e-05,
|
| 79 |
+
"loss": 2.8114,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 1.21,
|
| 84 |
+
"learning_rate": 4.3967118707040685e-05,
|
| 85 |
+
"loss": 2.8242,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 1.3,
|
| 90 |
+
"learning_rate": 4.350269366524243e-05,
|
| 91 |
+
"loss": 2.8101,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 1.39,
|
| 96 |
+
"learning_rate": 4.303826862344418e-05,
|
| 97 |
+
"loss": 2.8055,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 1.49,
|
| 102 |
+
"learning_rate": 4.2573843581645924e-05,
|
| 103 |
+
"loss": 2.821,
|
| 104 |
+
"step": 8000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 1.58,
|
| 108 |
+
"learning_rate": 4.210941853984767e-05,
|
| 109 |
+
"loss": 2.8286,
|
| 110 |
+
"step": 8500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 1.67,
|
| 114 |
+
"learning_rate": 4.1644993498049416e-05,
|
| 115 |
+
"loss": 2.8026,
|
| 116 |
+
"step": 9000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.76,
|
| 120 |
+
"learning_rate": 4.118056845625116e-05,
|
| 121 |
+
"loss": 2.8078,
|
| 122 |
+
"step": 9500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 1.86,
|
| 126 |
+
"learning_rate": 4.0716143414452915e-05,
|
| 127 |
+
"loss": 2.7948,
|
| 128 |
+
"step": 10000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 1.95,
|
| 132 |
+
"learning_rate": 4.0251718372654655e-05,
|
| 133 |
+
"loss": 2.8053,
|
| 134 |
+
"step": 10500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 2.04,
|
| 138 |
+
"learning_rate": 3.97872933308564e-05,
|
| 139 |
+
"loss": 2.6906,
|
| 140 |
+
"step": 11000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 2.14,
|
| 144 |
+
"learning_rate": 3.9323797139141746e-05,
|
| 145 |
+
"loss": 2.5819,
|
| 146 |
+
"step": 11500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 2.23,
|
| 150 |
+
"learning_rate": 3.8859372097343485e-05,
|
| 151 |
+
"loss": 2.5768,
|
| 152 |
+
"step": 12000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 2.32,
|
| 156 |
+
"learning_rate": 3.839587590562883e-05,
|
| 157 |
+
"loss": 2.597,
|
| 158 |
+
"step": 12500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 2.42,
|
| 162 |
+
"learning_rate": 3.7931450863830584e-05,
|
| 163 |
+
"loss": 2.5986,
|
| 164 |
+
"step": 13000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 2.51,
|
| 168 |
+
"learning_rate": 3.746702582203232e-05,
|
| 169 |
+
"loss": 2.6172,
|
| 170 |
+
"step": 13500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 2.6,
|
| 174 |
+
"learning_rate": 3.700260078023407e-05,
|
| 175 |
+
"loss": 2.5917,
|
| 176 |
+
"step": 14000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 2.69,
|
| 180 |
+
"learning_rate": 3.653817573843582e-05,
|
| 181 |
+
"loss": 2.6019,
|
| 182 |
+
"step": 14500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 2.79,
|
| 186 |
+
"learning_rate": 3.607375069663756e-05,
|
| 187 |
+
"loss": 2.598,
|
| 188 |
+
"step": 15000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 2.88,
|
| 192 |
+
"learning_rate": 3.560932565483931e-05,
|
| 193 |
+
"loss": 2.5893,
|
| 194 |
+
"step": 15500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 2.97,
|
| 198 |
+
"learning_rate": 3.514490061304106e-05,
|
| 199 |
+
"loss": 2.589,
|
| 200 |
+
"step": 16000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 3.07,
|
| 204 |
+
"learning_rate": 3.468047557124281e-05,
|
| 205 |
+
"loss": 2.4718,
|
| 206 |
+
"step": 16500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 3.16,
|
| 210 |
+
"learning_rate": 3.4216979379528146e-05,
|
| 211 |
+
"loss": 2.41,
|
| 212 |
+
"step": 17000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 3.25,
|
| 216 |
+
"learning_rate": 3.375255433772989e-05,
|
| 217 |
+
"loss": 2.3767,
|
| 218 |
+
"step": 17500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 3.34,
|
| 222 |
+
"learning_rate": 3.328812929593164e-05,
|
| 223 |
+
"loss": 2.3999,
|
| 224 |
+
"step": 18000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 3.44,
|
| 228 |
+
"learning_rate": 3.2823704254133384e-05,
|
| 229 |
+
"loss": 2.4211,
|
| 230 |
+
"step": 18500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 3.53,
|
| 234 |
+
"learning_rate": 3.236020806241873e-05,
|
| 235 |
+
"loss": 2.4202,
|
| 236 |
+
"step": 19000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 3.62,
|
| 240 |
+
"learning_rate": 3.1895783020620476e-05,
|
| 241 |
+
"loss": 2.4131,
|
| 242 |
+
"step": 19500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 3.72,
|
| 246 |
+
"learning_rate": 3.1431357978822215e-05,
|
| 247 |
+
"loss": 2.4155,
|
| 248 |
+
"step": 20000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 3.81,
|
| 252 |
+
"learning_rate": 3.096693293702397e-05,
|
| 253 |
+
"loss": 2.3968,
|
| 254 |
+
"step": 20500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 3.9,
|
| 258 |
+
"learning_rate": 3.050343674530931e-05,
|
| 259 |
+
"loss": 2.4189,
|
| 260 |
+
"step": 21000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 3.99,
|
| 264 |
+
"learning_rate": 3.0039011703511056e-05,
|
| 265 |
+
"loss": 2.406,
|
| 266 |
+
"step": 21500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 4.09,
|
| 270 |
+
"learning_rate": 2.95745866617128e-05,
|
| 271 |
+
"loss": 2.2482,
|
| 272 |
+
"step": 22000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 4.18,
|
| 276 |
+
"learning_rate": 2.911016161991455e-05,
|
| 277 |
+
"loss": 2.2432,
|
| 278 |
+
"step": 22500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 4.27,
|
| 282 |
+
"learning_rate": 2.8646665428199894e-05,
|
| 283 |
+
"loss": 2.2532,
|
| 284 |
+
"step": 23000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 4.37,
|
| 288 |
+
"learning_rate": 2.8182240386401637e-05,
|
| 289 |
+
"loss": 2.2649,
|
| 290 |
+
"step": 23500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 4.46,
|
| 294 |
+
"learning_rate": 2.7717815344603383e-05,
|
| 295 |
+
"loss": 2.2365,
|
| 296 |
+
"step": 24000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 4.55,
|
| 300 |
+
"learning_rate": 2.7253390302805126e-05,
|
| 301 |
+
"loss": 2.233,
|
| 302 |
+
"step": 24500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 4.64,
|
| 306 |
+
"learning_rate": 2.6788965261006875e-05,
|
| 307 |
+
"loss": 2.2569,
|
| 308 |
+
"step": 25000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 4.74,
|
| 312 |
+
"learning_rate": 2.632454021920862e-05,
|
| 313 |
+
"loss": 2.2609,
|
| 314 |
+
"step": 25500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 4.83,
|
| 318 |
+
"learning_rate": 2.5860115177410364e-05,
|
| 319 |
+
"loss": 2.256,
|
| 320 |
+
"step": 26000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 4.92,
|
| 324 |
+
"learning_rate": 2.5395690135612117e-05,
|
| 325 |
+
"loss": 2.2553,
|
| 326 |
+
"step": 26500
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 5.02,
|
| 330 |
+
"learning_rate": 2.4933122793981055e-05,
|
| 331 |
+
"loss": 2.1951,
|
| 332 |
+
"step": 27000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 5.11,
|
| 336 |
+
"learning_rate": 2.4468697752182798e-05,
|
| 337 |
+
"loss": 2.1061,
|
| 338 |
+
"step": 27500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 5.2,
|
| 342 |
+
"learning_rate": 2.4004272710384544e-05,
|
| 343 |
+
"loss": 2.0723,
|
| 344 |
+
"step": 28000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 5.29,
|
| 348 |
+
"learning_rate": 2.353984766858629e-05,
|
| 349 |
+
"loss": 2.0796,
|
| 350 |
+
"step": 28500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 5.39,
|
| 354 |
+
"learning_rate": 2.3075422626788036e-05,
|
| 355 |
+
"loss": 2.1206,
|
| 356 |
+
"step": 29000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 5.48,
|
| 360 |
+
"learning_rate": 2.261192643507338e-05,
|
| 361 |
+
"loss": 2.1276,
|
| 362 |
+
"step": 29500
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 5.57,
|
| 366 |
+
"learning_rate": 2.2147501393275128e-05,
|
| 367 |
+
"loss": 2.1108,
|
| 368 |
+
"step": 30000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 5.67,
|
| 372 |
+
"learning_rate": 2.168400520156047e-05,
|
| 373 |
+
"loss": 2.1241,
|
| 374 |
+
"step": 30500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 5.76,
|
| 378 |
+
"learning_rate": 2.1219580159762216e-05,
|
| 379 |
+
"loss": 2.1392,
|
| 380 |
+
"step": 31000
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 5.85,
|
| 384 |
+
"learning_rate": 2.0755155117963962e-05,
|
| 385 |
+
"loss": 2.1427,
|
| 386 |
+
"step": 31500
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 5.94,
|
| 390 |
+
"learning_rate": 2.0291658926249304e-05,
|
| 391 |
+
"loss": 2.0999,
|
| 392 |
+
"step": 32000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 6.04,
|
| 396 |
+
"learning_rate": 1.982723388445105e-05,
|
| 397 |
+
"loss": 2.0637,
|
| 398 |
+
"step": 32500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 6.13,
|
| 402 |
+
"learning_rate": 1.9362808842652796e-05,
|
| 403 |
+
"loss": 1.9977,
|
| 404 |
+
"step": 33000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 6.22,
|
| 408 |
+
"learning_rate": 1.8898383800854543e-05,
|
| 409 |
+
"loss": 1.98,
|
| 410 |
+
"step": 33500
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 6.32,
|
| 414 |
+
"learning_rate": 1.843395875905629e-05,
|
| 415 |
+
"loss": 1.9836,
|
| 416 |
+
"step": 34000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 6.41,
|
| 420 |
+
"learning_rate": 1.7969533717258035e-05,
|
| 421 |
+
"loss": 2.0058,
|
| 422 |
+
"step": 34500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 6.5,
|
| 426 |
+
"learning_rate": 1.750510867545978e-05,
|
| 427 |
+
"loss": 1.9843,
|
| 428 |
+
"step": 35000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 6.59,
|
| 432 |
+
"learning_rate": 1.7040683633661527e-05,
|
| 433 |
+
"loss": 1.9899,
|
| 434 |
+
"step": 35500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 6.69,
|
| 438 |
+
"learning_rate": 1.6576258591863277e-05,
|
| 439 |
+
"loss": 2.0094,
|
| 440 |
+
"step": 36000
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 6.78,
|
| 444 |
+
"learning_rate": 1.611183355006502e-05,
|
| 445 |
+
"loss": 2.0092,
|
| 446 |
+
"step": 36500
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 6.87,
|
| 450 |
+
"learning_rate": 1.5647408508266766e-05,
|
| 451 |
+
"loss": 2.0013,
|
| 452 |
+
"step": 37000
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 6.97,
|
| 456 |
+
"learning_rate": 1.5182983466468512e-05,
|
| 457 |
+
"loss": 2.0088,
|
| 458 |
+
"step": 37500
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 7.06,
|
| 462 |
+
"learning_rate": 1.4719487274753854e-05,
|
| 463 |
+
"loss": 1.9496,
|
| 464 |
+
"step": 38000
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 7.15,
|
| 468 |
+
"learning_rate": 1.4255062232955602e-05,
|
| 469 |
+
"loss": 1.8963,
|
| 470 |
+
"step": 38500
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 7.25,
|
| 474 |
+
"learning_rate": 1.379063719115735e-05,
|
| 475 |
+
"loss": 1.8895,
|
| 476 |
+
"step": 39000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 7.34,
|
| 480 |
+
"learning_rate": 1.3326212149359094e-05,
|
| 481 |
+
"loss": 1.9075,
|
| 482 |
+
"step": 39500
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 7.43,
|
| 486 |
+
"learning_rate": 1.2862715957644436e-05,
|
| 487 |
+
"loss": 1.9191,
|
| 488 |
+
"step": 40000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 7.52,
|
| 492 |
+
"learning_rate": 1.2398290915846184e-05,
|
| 493 |
+
"loss": 1.9075,
|
| 494 |
+
"step": 40500
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 7.62,
|
| 498 |
+
"learning_rate": 1.193386587404793e-05,
|
| 499 |
+
"loss": 1.9071,
|
| 500 |
+
"step": 41000
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 7.71,
|
| 504 |
+
"learning_rate": 1.1469440832249676e-05,
|
| 505 |
+
"loss": 1.9168,
|
| 506 |
+
"step": 41500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 7.8,
|
| 510 |
+
"learning_rate": 1.100501579045142e-05,
|
| 511 |
+
"loss": 1.9142,
|
| 512 |
+
"step": 42000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 7.9,
|
| 516 |
+
"learning_rate": 1.0540590748653167e-05,
|
| 517 |
+
"loss": 1.8864,
|
| 518 |
+
"step": 42500
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 7.99,
|
| 522 |
+
"learning_rate": 1.0076165706854915e-05,
|
| 523 |
+
"loss": 1.9086,
|
| 524 |
+
"step": 43000
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 8.08,
|
| 528 |
+
"learning_rate": 9.61174066505666e-06,
|
| 529 |
+
"loss": 1.8377,
|
| 530 |
+
"step": 43500
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 8.17,
|
| 534 |
+
"learning_rate": 9.148244473342003e-06,
|
| 535 |
+
"loss": 1.8389,
|
| 536 |
+
"step": 44000
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 8.27,
|
| 540 |
+
"learning_rate": 8.683819431543749e-06,
|
| 541 |
+
"loss": 1.8276,
|
| 542 |
+
"step": 44500
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 8.36,
|
| 546 |
+
"learning_rate": 8.219394389745495e-06,
|
| 547 |
+
"loss": 1.8399,
|
| 548 |
+
"step": 45000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 8.45,
|
| 552 |
+
"learning_rate": 7.754969347947241e-06,
|
| 553 |
+
"loss": 1.8389,
|
| 554 |
+
"step": 45500
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 8.55,
|
| 558 |
+
"learning_rate": 7.29240200631618e-06,
|
| 559 |
+
"loss": 1.8308,
|
| 560 |
+
"step": 46000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 8.64,
|
| 564 |
+
"learning_rate": 6.827976964517926e-06,
|
| 565 |
+
"loss": 1.8329,
|
| 566 |
+
"step": 46500
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 8.73,
|
| 570 |
+
"learning_rate": 6.363551922719674e-06,
|
| 571 |
+
"loss": 1.8351,
|
| 572 |
+
"step": 47000
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 8.82,
|
| 576 |
+
"learning_rate": 5.8991268809214195e-06,
|
| 577 |
+
"loss": 1.8573,
|
| 578 |
+
"step": 47500
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 8.92,
|
| 582 |
+
"learning_rate": 5.434701839123166e-06,
|
| 583 |
+
"loss": 1.8221,
|
| 584 |
+
"step": 48000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 9.01,
|
| 588 |
+
"learning_rate": 4.9712056474085085e-06,
|
| 589 |
+
"loss": 1.8384,
|
| 590 |
+
"step": 48500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 9.1,
|
| 594 |
+
"learning_rate": 4.506780605610255e-06,
|
| 595 |
+
"loss": 1.7756,
|
| 596 |
+
"step": 49000
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 9.2,
|
| 600 |
+
"learning_rate": 4.042355563812001e-06,
|
| 601 |
+
"loss": 1.7839,
|
| 602 |
+
"step": 49500
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 9.29,
|
| 606 |
+
"learning_rate": 3.577930522013747e-06,
|
| 607 |
+
"loss": 1.7839,
|
| 608 |
+
"step": 50000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 9.38,
|
| 612 |
+
"learning_rate": 3.113505480215493e-06,
|
| 613 |
+
"loss": 1.8112,
|
| 614 |
+
"step": 50500
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 9.47,
|
| 618 |
+
"learning_rate": 2.6490804384172397e-06,
|
| 619 |
+
"loss": 1.8078,
|
| 620 |
+
"step": 51000
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 9.57,
|
| 624 |
+
"learning_rate": 2.184655396618986e-06,
|
| 625 |
+
"loss": 1.7745,
|
| 626 |
+
"step": 51500
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 9.66,
|
| 630 |
+
"learning_rate": 1.720230354820732e-06,
|
| 631 |
+
"loss": 1.7955,
|
| 632 |
+
"step": 52000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 9.75,
|
| 636 |
+
"learning_rate": 1.2558053130224781e-06,
|
| 637 |
+
"loss": 1.7978,
|
| 638 |
+
"step": 52500
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 9.85,
|
| 642 |
+
"learning_rate": 7.923091213078211e-07,
|
| 643 |
+
"loss": 1.8062,
|
| 644 |
+
"step": 53000
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 9.94,
|
| 648 |
+
"learning_rate": 3.2788407950956715e-07,
|
| 649 |
+
"loss": 1.7702,
|
| 650 |
+
"step": 53500
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 10.0,
|
| 654 |
+
"step": 53830,
|
| 655 |
+
"total_flos": 9.998391677288448e+16,
|
| 656 |
+
"train_loss": 2.279561182284652,
|
| 657 |
+
"train_runtime": 34779.9602,
|
| 658 |
+
"train_samples_per_second": 1.548,
|
| 659 |
+
"train_steps_per_second": 1.548
|
| 660 |
+
}
|
| 661 |
+
],
|
| 662 |
+
"max_steps": 53830,
|
| 663 |
+
"num_train_epochs": 10,
|
| 664 |
+
"total_flos": 9.998391677288448e+16,
|
| 665 |
+
"trial_name": null,
|
| 666 |
+
"trial_params": null
|
| 667 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:58dc3211ef5caee8bd7fbaf70f4d5e5576c14e4d360888d0bff9d2fe95dcea11
|
| 3 |
+
size 2671
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|