Upload 14 files
Browse files- config.json +31 -0
- generation_config.json +6 -0
- merges.txt +0 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scaler.pt +3 -0
- scheduler.pt +3 -0
- special_tokens_map.json +5 -0
- tokenizer.json +0 -0
- tokenizer_config.json +9 -0
- trainer_state.json +601 -0
- training_args.bin +3 -0
- vocab.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"activation_function": "gelu_new",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPT2LMHeadModel"
|
| 5 |
+
],
|
| 6 |
+
"attn_pdrop": 0.1,
|
| 7 |
+
"bos_token_id": 50256,
|
| 8 |
+
"embd_pdrop": 0.1,
|
| 9 |
+
"eos_token_id": 50256,
|
| 10 |
+
"initializer_range": 0.02,
|
| 11 |
+
"layer_norm_epsilon": 1e-05,
|
| 12 |
+
"model_type": "gpt2",
|
| 13 |
+
"n_embd": 768,
|
| 14 |
+
"n_head": 12,
|
| 15 |
+
"n_inner": null,
|
| 16 |
+
"n_layer": 12,
|
| 17 |
+
"n_positions": 1024,
|
| 18 |
+
"reorder_and_upcast_attn": false,
|
| 19 |
+
"resid_pdrop": 0.1,
|
| 20 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 21 |
+
"scale_attn_weights": true,
|
| 22 |
+
"summary_activation": null,
|
| 23 |
+
"summary_first_dropout": 0.1,
|
| 24 |
+
"summary_proj_to_labels": true,
|
| 25 |
+
"summary_type": "cls_index",
|
| 26 |
+
"summary_use_proj": true,
|
| 27 |
+
"torch_dtype": "float32",
|
| 28 |
+
"transformers_version": "4.29.0.dev0",
|
| 29 |
+
"use_cache": true,
|
| 30 |
+
"vocab_size": 50257
|
| 31 |
+
}
|
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": "4.29.0.dev0"
|
| 6 |
+
}
|
merges.txt
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bb50bb39696a8486a40ca9006ddb25ad091c15a2d0118f4baa91c1050e4d8e73
|
| 3 |
+
size 995604017
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:6d6e23c8533f24c32a7ed86ca5c04408c30d2ec560a556a6f08df26c052ab7ab
|
| 3 |
+
size 510396521
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f96808c9938a930525715eb2fddd2dd78e438d0ba0a19902f6c7f705dba577d3
|
| 3 |
+
size 14503
|
scaler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:86807d5de345e31d4942dd66d818a90bc4bb3d823b548bbd617da213dcc32265
|
| 3 |
+
size 559
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9a7a7f9f487910b01e20fd10a9578c5221f6c4419e20f86d80e4f497963fc285
|
| 3 |
+
size 623
|
special_tokens_map.json
ADDED
|
@@ -0,0 +1,5 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"bos_token": "<|endoftext|>",
|
| 3 |
+
"eos_token": "<|endoftext|>",
|
| 4 |
+
"unk_token": "<|endoftext|>"
|
| 5 |
+
}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 8 |
+
"unk_token": "<|endoftext|>"
|
| 9 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,601 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 4.17286491394043,
|
| 3 |
+
"best_model_checkpoint": "/home/bel3/portfolio-repo-TheBguy87/final project/model_folder/checkpoint-19500",
|
| 4 |
+
"epoch": 2.9601518026565463,
|
| 5 |
+
"global_step": 19500,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.08,
|
| 12 |
+
"learning_rate": 4.873488183796367e-05,
|
| 13 |
+
"loss": 6.5463,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.08,
|
| 18 |
+
"eval_accuracy": 0.21852576647097194,
|
| 19 |
+
"eval_loss": 5.444242477416992,
|
| 20 |
+
"eval_runtime": 27.5824,
|
| 21 |
+
"eval_samples_per_second": 51.12,
|
| 22 |
+
"eval_steps_per_second": 6.417,
|
| 23 |
+
"step": 500
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.15,
|
| 27 |
+
"learning_rate": 4.746976367592733e-05,
|
| 28 |
+
"loss": 5.7814,
|
| 29 |
+
"step": 1000
|
| 30 |
+
},
|
| 31 |
+
{
|
| 32 |
+
"epoch": 0.15,
|
| 33 |
+
"eval_accuracy": 0.2316206575897628,
|
| 34 |
+
"eval_loss": 5.191591739654541,
|
| 35 |
+
"eval_runtime": 27.5501,
|
| 36 |
+
"eval_samples_per_second": 51.179,
|
| 37 |
+
"eval_steps_per_second": 6.425,
|
| 38 |
+
"step": 1000
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.23,
|
| 42 |
+
"learning_rate": 4.6204645513891e-05,
|
| 43 |
+
"loss": 5.5475,
|
| 44 |
+
"step": 1500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.23,
|
| 48 |
+
"eval_accuracy": 0.2393984816310669,
|
| 49 |
+
"eval_loss": 5.062835693359375,
|
| 50 |
+
"eval_runtime": 27.5707,
|
| 51 |
+
"eval_samples_per_second": 51.141,
|
| 52 |
+
"eval_steps_per_second": 6.42,
|
| 53 |
+
"step": 1500
|
| 54 |
+
},
|
| 55 |
+
{
|
| 56 |
+
"epoch": 0.3,
|
| 57 |
+
"learning_rate": 4.493952735185467e-05,
|
| 58 |
+
"loss": 5.4202,
|
| 59 |
+
"step": 2000
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.3,
|
| 63 |
+
"eval_accuracy": 0.24114169130199442,
|
| 64 |
+
"eval_loss": 4.970830917358398,
|
| 65 |
+
"eval_runtime": 27.5621,
|
| 66 |
+
"eval_samples_per_second": 51.157,
|
| 67 |
+
"eval_steps_per_second": 6.422,
|
| 68 |
+
"step": 2000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.38,
|
| 72 |
+
"learning_rate": 4.367440918981833e-05,
|
| 73 |
+
"loss": 5.3562,
|
| 74 |
+
"step": 2500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.38,
|
| 78 |
+
"eval_accuracy": 0.24209795839058446,
|
| 79 |
+
"eval_loss": 4.920139312744141,
|
| 80 |
+
"eval_runtime": 27.5764,
|
| 81 |
+
"eval_samples_per_second": 51.131,
|
| 82 |
+
"eval_steps_per_second": 6.419,
|
| 83 |
+
"step": 2500
|
| 84 |
+
},
|
| 85 |
+
{
|
| 86 |
+
"epoch": 0.46,
|
| 87 |
+
"learning_rate": 4.240929102778199e-05,
|
| 88 |
+
"loss": 5.2881,
|
| 89 |
+
"step": 3000
|
| 90 |
+
},
|
| 91 |
+
{
|
| 92 |
+
"epoch": 0.46,
|
| 93 |
+
"eval_accuracy": 0.2436371459105356,
|
| 94 |
+
"eval_loss": 4.872959136962891,
|
| 95 |
+
"eval_runtime": 27.5692,
|
| 96 |
+
"eval_samples_per_second": 51.144,
|
| 97 |
+
"eval_steps_per_second": 6.42,
|
| 98 |
+
"step": 3000
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.53,
|
| 102 |
+
"learning_rate": 4.114417286574566e-05,
|
| 103 |
+
"loss": 5.2296,
|
| 104 |
+
"step": 3500
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.53,
|
| 108 |
+
"eval_accuracy": 0.2506516217679144,
|
| 109 |
+
"eval_loss": 4.810730457305908,
|
| 110 |
+
"eval_runtime": 27.5767,
|
| 111 |
+
"eval_samples_per_second": 51.13,
|
| 112 |
+
"eval_steps_per_second": 6.418,
|
| 113 |
+
"step": 3500
|
| 114 |
+
},
|
| 115 |
+
{
|
| 116 |
+
"epoch": 0.61,
|
| 117 |
+
"learning_rate": 3.987905470370933e-05,
|
| 118 |
+
"loss": 5.1578,
|
| 119 |
+
"step": 4000
|
| 120 |
+
},
|
| 121 |
+
{
|
| 122 |
+
"epoch": 0.61,
|
| 123 |
+
"eval_accuracy": 0.2533899598895227,
|
| 124 |
+
"eval_loss": 4.767050743103027,
|
| 125 |
+
"eval_runtime": 27.5725,
|
| 126 |
+
"eval_samples_per_second": 51.138,
|
| 127 |
+
"eval_steps_per_second": 6.419,
|
| 128 |
+
"step": 4000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.68,
|
| 132 |
+
"learning_rate": 3.8613936541672994e-05,
|
| 133 |
+
"loss": 5.0779,
|
| 134 |
+
"step": 4500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.68,
|
| 138 |
+
"eval_accuracy": 0.2565418939362396,
|
| 139 |
+
"eval_loss": 4.725719451904297,
|
| 140 |
+
"eval_runtime": 27.5923,
|
| 141 |
+
"eval_samples_per_second": 51.101,
|
| 142 |
+
"eval_steps_per_second": 6.415,
|
| 143 |
+
"step": 4500
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.76,
|
| 147 |
+
"learning_rate": 3.734881837963666e-05,
|
| 148 |
+
"loss": 5.0276,
|
| 149 |
+
"step": 5000
|
| 150 |
+
},
|
| 151 |
+
{
|
| 152 |
+
"epoch": 0.76,
|
| 153 |
+
"eval_accuracy": 0.2623502796630165,
|
| 154 |
+
"eval_loss": 4.667919635772705,
|
| 155 |
+
"eval_runtime": 27.5734,
|
| 156 |
+
"eval_samples_per_second": 51.136,
|
| 157 |
+
"eval_steps_per_second": 6.419,
|
| 158 |
+
"step": 5000
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.83,
|
| 162 |
+
"learning_rate": 3.6083700217600326e-05,
|
| 163 |
+
"loss": 5.0059,
|
| 164 |
+
"step": 5500
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.83,
|
| 168 |
+
"eval_accuracy": 0.2663904734146646,
|
| 169 |
+
"eval_loss": 4.631860256195068,
|
| 170 |
+
"eval_runtime": 27.5657,
|
| 171 |
+
"eval_samples_per_second": 51.151,
|
| 172 |
+
"eval_steps_per_second": 6.421,
|
| 173 |
+
"step": 5500
|
| 174 |
+
},
|
| 175 |
+
{
|
| 176 |
+
"epoch": 0.91,
|
| 177 |
+
"learning_rate": 3.481858205556399e-05,
|
| 178 |
+
"loss": 4.9264,
|
| 179 |
+
"step": 6000
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.91,
|
| 183 |
+
"eval_accuracy": 0.2692717658325353,
|
| 184 |
+
"eval_loss": 4.584375858306885,
|
| 185 |
+
"eval_runtime": 27.5605,
|
| 186 |
+
"eval_samples_per_second": 51.16,
|
| 187 |
+
"eval_steps_per_second": 6.422,
|
| 188 |
+
"step": 6000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.99,
|
| 192 |
+
"learning_rate": 3.355346389352766e-05,
|
| 193 |
+
"loss": 4.9049,
|
| 194 |
+
"step": 6500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.99,
|
| 198 |
+
"eval_accuracy": 0.272461173335554,
|
| 199 |
+
"eval_loss": 4.553061485290527,
|
| 200 |
+
"eval_runtime": 27.5742,
|
| 201 |
+
"eval_samples_per_second": 51.135,
|
| 202 |
+
"eval_steps_per_second": 6.419,
|
| 203 |
+
"step": 6500
|
| 204 |
+
},
|
| 205 |
+
{
|
| 206 |
+
"epoch": 1.06,
|
| 207 |
+
"learning_rate": 3.228834573149132e-05,
|
| 208 |
+
"loss": 4.8232,
|
| 209 |
+
"step": 7000
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 1.06,
|
| 213 |
+
"eval_accuracy": 0.27501491998723127,
|
| 214 |
+
"eval_loss": 4.517702579498291,
|
| 215 |
+
"eval_runtime": 27.5623,
|
| 216 |
+
"eval_samples_per_second": 51.157,
|
| 217 |
+
"eval_steps_per_second": 6.422,
|
| 218 |
+
"step": 7000
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 1.14,
|
| 222 |
+
"learning_rate": 3.102322756945499e-05,
|
| 223 |
+
"loss": 4.751,
|
| 224 |
+
"step": 7500
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 1.14,
|
| 228 |
+
"eval_accuracy": 0.2793854353166507,
|
| 229 |
+
"eval_loss": 4.4822540283203125,
|
| 230 |
+
"eval_runtime": 27.5718,
|
| 231 |
+
"eval_samples_per_second": 51.139,
|
| 232 |
+
"eval_steps_per_second": 6.42,
|
| 233 |
+
"step": 7500
|
| 234 |
+
},
|
| 235 |
+
{
|
| 236 |
+
"epoch": 1.21,
|
| 237 |
+
"learning_rate": 2.9758109407418657e-05,
|
| 238 |
+
"loss": 4.7553,
|
| 239 |
+
"step": 8000
|
| 240 |
+
},
|
| 241 |
+
{
|
| 242 |
+
"epoch": 1.21,
|
| 243 |
+
"eval_accuracy": 0.2807802806345505,
|
| 244 |
+
"eval_loss": 4.459610939025879,
|
| 245 |
+
"eval_runtime": 27.5709,
|
| 246 |
+
"eval_samples_per_second": 51.141,
|
| 247 |
+
"eval_steps_per_second": 6.42,
|
| 248 |
+
"step": 8000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 1.29,
|
| 252 |
+
"learning_rate": 2.8495521481706393e-05,
|
| 253 |
+
"loss": 4.7213,
|
| 254 |
+
"step": 8500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 1.29,
|
| 258 |
+
"eval_accuracy": 0.28297872340425534,
|
| 259 |
+
"eval_loss": 4.435904026031494,
|
| 260 |
+
"eval_runtime": 27.5742,
|
| 261 |
+
"eval_samples_per_second": 51.135,
|
| 262 |
+
"eval_steps_per_second": 6.419,
|
| 263 |
+
"step": 8500
|
| 264 |
+
},
|
| 265 |
+
{
|
| 266 |
+
"epoch": 1.37,
|
| 267 |
+
"learning_rate": 2.7230403319670056e-05,
|
| 268 |
+
"loss": 4.6646,
|
| 269 |
+
"step": 9000
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 1.37,
|
| 273 |
+
"eval_accuracy": 0.28641656604349697,
|
| 274 |
+
"eval_loss": 4.399264335632324,
|
| 275 |
+
"eval_runtime": 27.5882,
|
| 276 |
+
"eval_samples_per_second": 51.109,
|
| 277 |
+
"eval_steps_per_second": 6.416,
|
| 278 |
+
"step": 9000
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 1.44,
|
| 282 |
+
"learning_rate": 2.597034563028187e-05,
|
| 283 |
+
"loss": 4.6709,
|
| 284 |
+
"step": 9500
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 1.44,
|
| 288 |
+
"eval_accuracy": 0.2876531901014559,
|
| 289 |
+
"eval_loss": 4.388927936553955,
|
| 290 |
+
"eval_runtime": 27.5803,
|
| 291 |
+
"eval_samples_per_second": 51.123,
|
| 292 |
+
"eval_steps_per_second": 6.418,
|
| 293 |
+
"step": 9500
|
| 294 |
+
},
|
| 295 |
+
{
|
| 296 |
+
"epoch": 1.52,
|
| 297 |
+
"learning_rate": 2.4705227468245535e-05,
|
| 298 |
+
"loss": 4.6179,
|
| 299 |
+
"step": 10000
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 1.52,
|
| 303 |
+
"eval_accuracy": 0.2888079277178665,
|
| 304 |
+
"eval_loss": 4.365257263183594,
|
| 305 |
+
"eval_runtime": 27.58,
|
| 306 |
+
"eval_samples_per_second": 51.124,
|
| 307 |
+
"eval_steps_per_second": 6.418,
|
| 308 |
+
"step": 10000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 1.59,
|
| 312 |
+
"learning_rate": 2.34401093062092e-05,
|
| 313 |
+
"loss": 4.5984,
|
| 314 |
+
"step": 10500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 1.59,
|
| 318 |
+
"eval_accuracy": 0.2886746887621268,
|
| 319 |
+
"eval_loss": 4.34913444519043,
|
| 320 |
+
"eval_runtime": 27.5823,
|
| 321 |
+
"eval_samples_per_second": 51.12,
|
| 322 |
+
"eval_steps_per_second": 6.417,
|
| 323 |
+
"step": 10500
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 1.67,
|
| 327 |
+
"learning_rate": 2.2174991144172867e-05,
|
| 328 |
+
"loss": 4.5868,
|
| 329 |
+
"step": 11000
|
| 330 |
+
},
|
| 331 |
+
{
|
| 332 |
+
"epoch": 1.67,
|
| 333 |
+
"eval_accuracy": 0.29093697519812356,
|
| 334 |
+
"eval_loss": 4.338951587677002,
|
| 335 |
+
"eval_runtime": 27.5873,
|
| 336 |
+
"eval_samples_per_second": 51.11,
|
| 337 |
+
"eval_steps_per_second": 6.416,
|
| 338 |
+
"step": 11000
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 1.75,
|
| 342 |
+
"learning_rate": 2.0909872982136533e-05,
|
| 343 |
+
"loss": 4.5543,
|
| 344 |
+
"step": 11500
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 1.75,
|
| 348 |
+
"eval_accuracy": 0.29370307143551094,
|
| 349 |
+
"eval_loss": 4.3105316162109375,
|
| 350 |
+
"eval_runtime": 27.5819,
|
| 351 |
+
"eval_samples_per_second": 51.12,
|
| 352 |
+
"eval_steps_per_second": 6.417,
|
| 353 |
+
"step": 11500
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 1.82,
|
| 357 |
+
"learning_rate": 1.9644754820100196e-05,
|
| 358 |
+
"loss": 4.5493,
|
| 359 |
+
"step": 12000
|
| 360 |
+
},
|
| 361 |
+
{
|
| 362 |
+
"epoch": 1.82,
|
| 363 |
+
"eval_accuracy": 0.29424019097583654,
|
| 364 |
+
"eval_loss": 4.301023483276367,
|
| 365 |
+
"eval_runtime": 27.5881,
|
| 366 |
+
"eval_samples_per_second": 51.109,
|
| 367 |
+
"eval_steps_per_second": 6.416,
|
| 368 |
+
"step": 12000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 1.9,
|
| 372 |
+
"learning_rate": 1.8379636658063865e-05,
|
| 373 |
+
"loss": 4.5038,
|
| 374 |
+
"step": 12500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 1.9,
|
| 378 |
+
"eval_accuracy": 0.294011186520659,
|
| 379 |
+
"eval_loss": 4.281564235687256,
|
| 380 |
+
"eval_runtime": 27.5889,
|
| 381 |
+
"eval_samples_per_second": 51.107,
|
| 382 |
+
"eval_steps_per_second": 6.416,
|
| 383 |
+
"step": 12500
|
| 384 |
+
},
|
| 385 |
+
{
|
| 386 |
+
"epoch": 1.97,
|
| 387 |
+
"learning_rate": 1.7114518496027528e-05,
|
| 388 |
+
"loss": 4.4836,
|
| 389 |
+
"step": 13000
|
| 390 |
+
},
|
| 391 |
+
{
|
| 392 |
+
"epoch": 1.97,
|
| 393 |
+
"eval_accuracy": 0.29663155265020613,
|
| 394 |
+
"eval_loss": 4.268117427825928,
|
| 395 |
+
"eval_runtime": 27.5869,
|
| 396 |
+
"eval_samples_per_second": 51.111,
|
| 397 |
+
"eval_steps_per_second": 6.416,
|
| 398 |
+
"step": 13000
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 2.05,
|
| 402 |
+
"learning_rate": 1.5849400333991198e-05,
|
| 403 |
+
"loss": 4.4394,
|
| 404 |
+
"step": 13500
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 2.05,
|
| 408 |
+
"eval_accuracy": 0.2960416926899002,
|
| 409 |
+
"eval_loss": 4.262997150421143,
|
| 410 |
+
"eval_runtime": 27.5906,
|
| 411 |
+
"eval_samples_per_second": 51.104,
|
| 412 |
+
"eval_steps_per_second": 6.415,
|
| 413 |
+
"step": 13500
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 2.13,
|
| 417 |
+
"learning_rate": 1.458428217195486e-05,
|
| 418 |
+
"loss": 4.4429,
|
| 419 |
+
"step": 14000
|
| 420 |
+
},
|
| 421 |
+
{
|
| 422 |
+
"epoch": 2.13,
|
| 423 |
+
"eval_accuracy": 0.2994989660101872,
|
| 424 |
+
"eval_loss": 4.247574329376221,
|
| 425 |
+
"eval_runtime": 27.5874,
|
| 426 |
+
"eval_samples_per_second": 51.11,
|
| 427 |
+
"eval_steps_per_second": 6.416,
|
| 428 |
+
"step": 14000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 2.2,
|
| 432 |
+
"learning_rate": 1.3319164009918528e-05,
|
| 433 |
+
"loss": 4.4404,
|
| 434 |
+
"step": 14500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 2.2,
|
| 438 |
+
"eval_accuracy": 0.3004260870772092,
|
| 439 |
+
"eval_loss": 4.236292839050293,
|
| 440 |
+
"eval_runtime": 27.9352,
|
| 441 |
+
"eval_samples_per_second": 50.474,
|
| 442 |
+
"eval_steps_per_second": 6.336,
|
| 443 |
+
"step": 14500
|
| 444 |
+
},
|
| 445 |
+
{
|
| 446 |
+
"epoch": 2.28,
|
| 447 |
+
"learning_rate": 1.2054045847882193e-05,
|
| 448 |
+
"loss": 4.4148,
|
| 449 |
+
"step": 15000
|
| 450 |
+
},
|
| 451 |
+
{
|
| 452 |
+
"epoch": 2.28,
|
| 453 |
+
"eval_accuracy": 0.3013920695063219,
|
| 454 |
+
"eval_loss": 4.226977825164795,
|
| 455 |
+
"eval_runtime": 27.9368,
|
| 456 |
+
"eval_samples_per_second": 50.471,
|
| 457 |
+
"eval_steps_per_second": 6.336,
|
| 458 |
+
"step": 15000
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 2.35,
|
| 462 |
+
"learning_rate": 1.0791457922169932e-05,
|
| 463 |
+
"loss": 4.3944,
|
| 464 |
+
"step": 15500
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 2.35,
|
| 468 |
+
"eval_accuracy": 0.3011366948411542,
|
| 469 |
+
"eval_loss": 4.218454360961914,
|
| 470 |
+
"eval_runtime": 28.163,
|
| 471 |
+
"eval_samples_per_second": 50.066,
|
| 472 |
+
"eval_steps_per_second": 6.285,
|
| 473 |
+
"step": 15500
|
| 474 |
+
},
|
| 475 |
+
{
|
| 476 |
+
"epoch": 2.43,
|
| 477 |
+
"learning_rate": 9.526339760133597e-06,
|
| 478 |
+
"loss": 4.3942,
|
| 479 |
+
"step": 16000
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 2.43,
|
| 483 |
+
"eval_accuracy": 0.3020194029229296,
|
| 484 |
+
"eval_loss": 4.213292121887207,
|
| 485 |
+
"eval_runtime": 28.5992,
|
| 486 |
+
"eval_samples_per_second": 49.302,
|
| 487 |
+
"eval_steps_per_second": 6.189,
|
| 488 |
+
"step": 16000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 2.5,
|
| 492 |
+
"learning_rate": 8.261221598097263e-06,
|
| 493 |
+
"loss": 4.3789,
|
| 494 |
+
"step": 16500
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 2.5,
|
| 498 |
+
"eval_accuracy": 0.3029826095404644,
|
| 499 |
+
"eval_loss": 4.205790996551514,
|
| 500 |
+
"eval_runtime": 28.126,
|
| 501 |
+
"eval_samples_per_second": 50.131,
|
| 502 |
+
"eval_steps_per_second": 6.293,
|
| 503 |
+
"step": 16500
|
| 504 |
+
},
|
| 505 |
+
{
|
| 506 |
+
"epoch": 2.58,
|
| 507 |
+
"learning_rate": 6.996103436060929e-06,
|
| 508 |
+
"loss": 4.3444,
|
| 509 |
+
"step": 17000
|
| 510 |
+
},
|
| 511 |
+
{
|
| 512 |
+
"epoch": 2.58,
|
| 513 |
+
"eval_accuracy": 0.3037764916517467,
|
| 514 |
+
"eval_loss": 4.199104309082031,
|
| 515 |
+
"eval_runtime": 27.8045,
|
| 516 |
+
"eval_samples_per_second": 50.711,
|
| 517 |
+
"eval_steps_per_second": 6.366,
|
| 518 |
+
"step": 17000
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 2.66,
|
| 522 |
+
"learning_rate": 5.733515510348667e-06,
|
| 523 |
+
"loss": 4.3503,
|
| 524 |
+
"step": 17500
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 2.66,
|
| 528 |
+
"eval_accuracy": 0.3041026495121511,
|
| 529 |
+
"eval_loss": 4.189632415771484,
|
| 530 |
+
"eval_runtime": 28.554,
|
| 531 |
+
"eval_samples_per_second": 49.38,
|
| 532 |
+
"eval_steps_per_second": 6.199,
|
| 533 |
+
"step": 17500
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 2.73,
|
| 537 |
+
"learning_rate": 4.468397348312332e-06,
|
| 538 |
+
"loss": 4.3445,
|
| 539 |
+
"step": 18000
|
| 540 |
+
},
|
| 541 |
+
{
|
| 542 |
+
"epoch": 2.73,
|
| 543 |
+
"eval_accuracy": 0.30501727942707246,
|
| 544 |
+
"eval_loss": 4.184421062469482,
|
| 545 |
+
"eval_runtime": 27.7494,
|
| 546 |
+
"eval_samples_per_second": 50.812,
|
| 547 |
+
"eval_steps_per_second": 6.379,
|
| 548 |
+
"step": 18000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 2.81,
|
| 552 |
+
"learning_rate": 3.2058094226000707e-06,
|
| 553 |
+
"loss": 4.3081,
|
| 554 |
+
"step": 18500
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 2.81,
|
| 558 |
+
"eval_accuracy": 0.3053531526279996,
|
| 559 |
+
"eval_loss": 4.178225994110107,
|
| 560 |
+
"eval_runtime": 27.5642,
|
| 561 |
+
"eval_samples_per_second": 51.153,
|
| 562 |
+
"eval_steps_per_second": 6.421,
|
| 563 |
+
"step": 18500
|
| 564 |
+
},
|
| 565 |
+
{
|
| 566 |
+
"epoch": 2.88,
|
| 567 |
+
"learning_rate": 1.9406912605637364e-06,
|
| 568 |
+
"loss": 4.3215,
|
| 569 |
+
"step": 19000
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 2.88,
|
| 573 |
+
"eval_accuracy": 0.30591109075515954,
|
| 574 |
+
"eval_loss": 4.176368713378906,
|
| 575 |
+
"eval_runtime": 27.554,
|
| 576 |
+
"eval_samples_per_second": 51.172,
|
| 577 |
+
"eval_steps_per_second": 6.424,
|
| 578 |
+
"step": 19000
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 2.96,
|
| 582 |
+
"learning_rate": 6.755730985274025e-07,
|
| 583 |
+
"loss": 4.3413,
|
| 584 |
+
"step": 19500
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 2.96,
|
| 588 |
+
"eval_accuracy": 0.30599714091407476,
|
| 589 |
+
"eval_loss": 4.17286491394043,
|
| 590 |
+
"eval_runtime": 27.9537,
|
| 591 |
+
"eval_samples_per_second": 50.441,
|
| 592 |
+
"eval_steps_per_second": 6.332,
|
| 593 |
+
"step": 19500
|
| 594 |
+
}
|
| 595 |
+
],
|
| 596 |
+
"max_steps": 19761,
|
| 597 |
+
"num_train_epochs": 3,
|
| 598 |
+
"total_flos": 2.0380255911936e+16,
|
| 599 |
+
"trial_name": null,
|
| 600 |
+
"trial_params": null
|
| 601 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e0524b826d7bcd879ed7d8eab0cc57c477df4c6d4ee1a9576bf2650ce6252d36
|
| 3 |
+
size 3951
|
vocab.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|