Upload folder using huggingface_hub
Browse files- config.json +48 -0
- generation_config.json +9 -0
- model.safetensors +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- tokenizer.json +0 -0
- tokenizer_config.json +14 -0
- trainer_state.json +524 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,48 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_num_labels": 1,
|
| 3 |
+
"activation_function": "gelu_new",
|
| 4 |
+
"add_cross_attention": false,
|
| 5 |
+
"architectures": [
|
| 6 |
+
"GPT2LMHeadModel"
|
| 7 |
+
],
|
| 8 |
+
"attn_pdrop": 0.1,
|
| 9 |
+
"bos_token_id": 50256,
|
| 10 |
+
"dtype": "float32",
|
| 11 |
+
"embd_pdrop": 0.1,
|
| 12 |
+
"eos_token_id": 50256,
|
| 13 |
+
"id2label": {
|
| 14 |
+
"0": "LABEL_0"
|
| 15 |
+
},
|
| 16 |
+
"initializer_range": 0.02,
|
| 17 |
+
"label2id": {
|
| 18 |
+
"LABEL_0": 0
|
| 19 |
+
},
|
| 20 |
+
"layer_norm_epsilon": 1e-05,
|
| 21 |
+
"model_type": "gpt2",
|
| 22 |
+
"n_ctx": 1024,
|
| 23 |
+
"n_embd": 768,
|
| 24 |
+
"n_head": 12,
|
| 25 |
+
"n_inner": null,
|
| 26 |
+
"n_layer": 6,
|
| 27 |
+
"n_positions": 1024,
|
| 28 |
+
"pad_token_id": 50256,
|
| 29 |
+
"reorder_and_upcast_attn": false,
|
| 30 |
+
"resid_pdrop": 0.1,
|
| 31 |
+
"scale_attn_by_inverse_layer_idx": false,
|
| 32 |
+
"scale_attn_weights": true,
|
| 33 |
+
"summary_activation": null,
|
| 34 |
+
"summary_first_dropout": 0.1,
|
| 35 |
+
"summary_proj_to_labels": true,
|
| 36 |
+
"summary_type": "cls_index",
|
| 37 |
+
"summary_use_proj": true,
|
| 38 |
+
"task_specific_params": {
|
| 39 |
+
"text-generation": {
|
| 40 |
+
"do_sample": true,
|
| 41 |
+
"max_length": 50
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"tie_word_embeddings": true,
|
| 45 |
+
"transformers_version": "5.0.0",
|
| 46 |
+
"use_cache": false,
|
| 47 |
+
"vocab_size": 50257
|
| 48 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,9 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 50256,
|
| 4 |
+
"eos_token_id": [
|
| 5 |
+
50256
|
| 6 |
+
],
|
| 7 |
+
"pad_token_id": 50256,
|
| 8 |
+
"transformers_version": "5.0.0"
|
| 9 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:033c075013a554bf84f4ebcd1dd5ee609e330f15d3bc9413983a62ff35c6e887
|
| 3 |
+
size 327657928
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2ad076501b6247f30f1b6a12d6e247a2dba08764bea7a13f2cc4caa827cf0e99
|
| 3 |
+
size 14645
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:523d8282e4a819e5e25c581fa82390b38e5026bf7911019b4261040f1bb0e371
|
| 3 |
+
size 1465
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,14 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"add_prefix_space": false,
|
| 3 |
+
"backend": "tokenizers",
|
| 4 |
+
"bos_token": "<|endoftext|>",
|
| 5 |
+
"clean_up_tokenization_spaces": false,
|
| 6 |
+
"eos_token": "<|endoftext|>",
|
| 7 |
+
"errors": "replace",
|
| 8 |
+
"is_local": false,
|
| 9 |
+
"model_max_length": 1024,
|
| 10 |
+
"model_specific_special_tokens": {},
|
| 11 |
+
"pad_token": "<|endoftext|>",
|
| 12 |
+
"tokenizer_class": "GPT2Tokenizer",
|
| 13 |
+
"unk_token": "<|endoftext|>"
|
| 14 |
+
}
|
trainer_state.json
ADDED
|
@@ -0,0 +1,524 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.4666666666666667,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 1400,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.006666666666666667,
|
| 14 |
+
"grad_norm": 3.3801944255828857,
|
| 15 |
+
"learning_rate": 9.5e-06,
|
| 16 |
+
"loss": 3.710173416137695,
|
| 17 |
+
"step": 20
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.013333333333333334,
|
| 21 |
+
"grad_norm": 2.4231245517730713,
|
| 22 |
+
"learning_rate": 1.9500000000000003e-05,
|
| 23 |
+
"loss": 3.2622753143310548,
|
| 24 |
+
"step": 40
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.02,
|
| 28 |
+
"grad_norm": 2.0525474548339844,
|
| 29 |
+
"learning_rate": 2.95e-05,
|
| 30 |
+
"loss": 2.287516975402832,
|
| 31 |
+
"step": 60
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.02666666666666667,
|
| 35 |
+
"grad_norm": 1.2226288318634033,
|
| 36 |
+
"learning_rate": 3.9500000000000005e-05,
|
| 37 |
+
"loss": 1.6385210037231446,
|
| 38 |
+
"step": 80
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.03333333333333333,
|
| 42 |
+
"grad_norm": 2.2779319286346436,
|
| 43 |
+
"learning_rate": 4.9500000000000004e-05,
|
| 44 |
+
"loss": 1.4153234481811523,
|
| 45 |
+
"step": 100
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.04,
|
| 49 |
+
"grad_norm": 1.5858861207962036,
|
| 50 |
+
"learning_rate": 4.9994704516344666e-05,
|
| 51 |
+
"loss": 1.4460668563842773,
|
| 52 |
+
"step": 120
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.04666666666666667,
|
| 56 |
+
"grad_norm": 1.251664161682129,
|
| 57 |
+
"learning_rate": 4.99776910887808e-05,
|
| 58 |
+
"loss": 1.5103114128112793,
|
| 59 |
+
"step": 140
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.05333333333333334,
|
| 63 |
+
"grad_norm": 2.4487648010253906,
|
| 64 |
+
"learning_rate": 4.9948953034032466e-05,
|
| 65 |
+
"loss": 1.3196531295776368,
|
| 66 |
+
"step": 160
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.06,
|
| 70 |
+
"grad_norm": 1.276078701019287,
|
| 71 |
+
"learning_rate": 4.990850384185648e-05,
|
| 72 |
+
"loss": 1.3049612045288086,
|
| 73 |
+
"step": 180
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.06666666666666667,
|
| 77 |
+
"grad_norm": 1.4780528545379639,
|
| 78 |
+
"learning_rate": 4.985636249926392e-05,
|
| 79 |
+
"loss": 1.164187526702881,
|
| 80 |
+
"step": 200
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.07333333333333333,
|
| 84 |
+
"grad_norm": 1.1906464099884033,
|
| 85 |
+
"learning_rate": 4.979255348160766e-05,
|
| 86 |
+
"loss": 1.418909740447998,
|
| 87 |
+
"step": 220
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.08,
|
| 91 |
+
"grad_norm": 1.3630824089050293,
|
| 92 |
+
"learning_rate": 4.9717106741093454e-05,
|
| 93 |
+
"loss": 1.0971190452575683,
|
| 94 |
+
"step": 240
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.08666666666666667,
|
| 98 |
+
"grad_norm": 1.3592742681503296,
|
| 99 |
+
"learning_rate": 4.96300576927203e-05,
|
| 100 |
+
"loss": 1.358650016784668,
|
| 101 |
+
"step": 260
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.09333333333333334,
|
| 105 |
+
"grad_norm": 1.215366244316101,
|
| 106 |
+
"learning_rate": 4.953144719765649e-05,
|
| 107 |
+
"loss": 1.3464448928833008,
|
| 108 |
+
"step": 280
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.1,
|
| 112 |
+
"grad_norm": 1.00632643699646,
|
| 113 |
+
"learning_rate": 4.942132154405919e-05,
|
| 114 |
+
"loss": 0.925452709197998,
|
| 115 |
+
"step": 300
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.10666666666666667,
|
| 119 |
+
"grad_norm": 1.3117424249649048,
|
| 120 |
+
"learning_rate": 4.9299732425346626e-05,
|
| 121 |
+
"loss": 1.4749733924865722,
|
| 122 |
+
"step": 320
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.11333333333333333,
|
| 126 |
+
"grad_norm": 1.153543472290039,
|
| 127 |
+
"learning_rate": 4.9166736915933e-05,
|
| 128 |
+
"loss": 1.339738368988037,
|
| 129 |
+
"step": 340
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.12,
|
| 133 |
+
"grad_norm": 1.147539496421814,
|
| 134 |
+
"learning_rate": 4.9022397444437495e-05,
|
| 135 |
+
"loss": 1.1870535850524901,
|
| 136 |
+
"step": 360
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.12666666666666668,
|
| 140 |
+
"grad_norm": 1.023755431175232,
|
| 141 |
+
"learning_rate": 4.886678176438004e-05,
|
| 142 |
+
"loss": 0.9843617439270019,
|
| 143 |
+
"step": 380
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.13333333333333333,
|
| 147 |
+
"grad_norm": 0.9747999906539917,
|
| 148 |
+
"learning_rate": 4.869996292237757e-05,
|
| 149 |
+
"loss": 0.9861419677734375,
|
| 150 |
+
"step": 400
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.14,
|
| 154 |
+
"grad_norm": 0.9171973466873169,
|
| 155 |
+
"learning_rate": 4.852201922385564e-05,
|
| 156 |
+
"loss": 1.001624870300293,
|
| 157 |
+
"step": 420
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.14666666666666667,
|
| 161 |
+
"grad_norm": 0.7914513945579529,
|
| 162 |
+
"learning_rate": 4.8333034196291516e-05,
|
| 163 |
+
"loss": 0.9351105690002441,
|
| 164 |
+
"step": 440
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.15333333333333332,
|
| 168 |
+
"grad_norm": 0.8947938680648804,
|
| 169 |
+
"learning_rate": 4.8133096550006096e-05,
|
| 170 |
+
"loss": 1.0424003601074219,
|
| 171 |
+
"step": 460
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.16,
|
| 175 |
+
"grad_norm": 0.7379234433174133,
|
| 176 |
+
"learning_rate": 4.7922300136522914e-05,
|
| 177 |
+
"loss": 0.9020758628845215,
|
| 178 |
+
"step": 480
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.16666666666666666,
|
| 182 |
+
"grad_norm": 0.720455527305603,
|
| 183 |
+
"learning_rate": 4.7700743904513836e-05,
|
| 184 |
+
"loss": 0.8897322654724121,
|
| 185 |
+
"step": 500
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.17333333333333334,
|
| 189 |
+
"grad_norm": 1.3379108905792236,
|
| 190 |
+
"learning_rate": 4.746853185335221e-05,
|
| 191 |
+
"loss": 0.9499833106994628,
|
| 192 |
+
"step": 520
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.18,
|
| 196 |
+
"grad_norm": 0.9506772756576538,
|
| 197 |
+
"learning_rate": 4.722577298429512e-05,
|
| 198 |
+
"loss": 1.4986319541931152,
|
| 199 |
+
"step": 540
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.18666666666666668,
|
| 203 |
+
"grad_norm": 1.5777019262313843,
|
| 204 |
+
"learning_rate": 4.697258124931776e-05,
|
| 205 |
+
"loss": 1.3104500770568848,
|
| 206 |
+
"step": 560
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.19333333333333333,
|
| 210 |
+
"grad_norm": 1.2068767547607422,
|
| 211 |
+
"learning_rate": 4.670907549762392e-05,
|
| 212 |
+
"loss": 0.9996993064880371,
|
| 213 |
+
"step": 580
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.2,
|
| 217 |
+
"grad_norm": 1.3350718021392822,
|
| 218 |
+
"learning_rate": 4.643537941985775e-05,
|
| 219 |
+
"loss": 1.0175305366516114,
|
| 220 |
+
"step": 600
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.20666666666666667,
|
| 224 |
+
"grad_norm": 1.7439497709274292,
|
| 225 |
+
"learning_rate": 4.6151621490042805e-05,
|
| 226 |
+
"loss": 1.0757975578308105,
|
| 227 |
+
"step": 620
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.21333333333333335,
|
| 231 |
+
"grad_norm": 1.2179644107818604,
|
| 232 |
+
"learning_rate": 4.5857934905275895e-05,
|
| 233 |
+
"loss": 1.1192784309387207,
|
| 234 |
+
"step": 640
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.22,
|
| 238 |
+
"grad_norm": 1.1882299184799194,
|
| 239 |
+
"learning_rate": 4.555445752320379e-05,
|
| 240 |
+
"loss": 0.9488797187805176,
|
| 241 |
+
"step": 660
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.22666666666666666,
|
| 245 |
+
"grad_norm": 1.0356523990631104,
|
| 246 |
+
"learning_rate": 4.524133179731231e-05,
|
| 247 |
+
"loss": 0.3407599449157715,
|
| 248 |
+
"step": 680
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.23333333333333334,
|
| 252 |
+
"grad_norm": 1.4150390625,
|
| 253 |
+
"learning_rate": 4.491870471005814e-05,
|
| 254 |
+
"loss": 0.9790312767028808,
|
| 255 |
+
"step": 700
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.24,
|
| 259 |
+
"grad_norm": 1.3073979616165161,
|
| 260 |
+
"learning_rate": 4.4586727703874606e-05,
|
| 261 |
+
"loss": 1.325801467895508,
|
| 262 |
+
"step": 720
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.24666666666666667,
|
| 266 |
+
"grad_norm": 1.2574576139450073,
|
| 267 |
+
"learning_rate": 4.424555661008406e-05,
|
| 268 |
+
"loss": 1.1854475021362305,
|
| 269 |
+
"step": 740
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.25333333333333335,
|
| 273 |
+
"grad_norm": 1.4099619388580322,
|
| 274 |
+
"learning_rate": 4.389535157574998e-05,
|
| 275 |
+
"loss": 1.0839804649353026,
|
| 276 |
+
"step": 760
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.26,
|
| 280 |
+
"grad_norm": 1.2009291648864746,
|
| 281 |
+
"learning_rate": 4.3536276988503366e-05,
|
| 282 |
+
"loss": 1.030893611907959,
|
| 283 |
+
"step": 780
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.26666666666666666,
|
| 287 |
+
"grad_norm": 1.0084775686264038,
|
| 288 |
+
"learning_rate": 4.316850139937849e-05,
|
| 289 |
+
"loss": 0.9282444000244141,
|
| 290 |
+
"step": 800
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.2733333333333333,
|
| 294 |
+
"grad_norm": 1.166970133781433,
|
| 295 |
+
"learning_rate": 4.27921974436944e-05,
|
| 296 |
+
"loss": 0.9386281967163086,
|
| 297 |
+
"step": 820
|
| 298 |
+
},
|
| 299 |
+
{
|
| 300 |
+
"epoch": 0.28,
|
| 301 |
+
"grad_norm": 1.1025112867355347,
|
| 302 |
+
"learning_rate": 4.24075417600192e-05,
|
| 303 |
+
"loss": 1.0037524223327636,
|
| 304 |
+
"step": 840
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 0.2866666666666667,
|
| 308 |
+
"grad_norm": 1.2833219766616821,
|
| 309 |
+
"learning_rate": 4.20147149072552e-05,
|
| 310 |
+
"loss": 0.7834305763244629,
|
| 311 |
+
"step": 860
|
| 312 |
+
},
|
| 313 |
+
{
|
| 314 |
+
"epoch": 0.29333333333333333,
|
| 315 |
+
"grad_norm": 1.5375336408615112,
|
| 316 |
+
"learning_rate": 4.1613901279883885e-05,
|
| 317 |
+
"loss": 1.0162606239318848,
|
| 318 |
+
"step": 880
|
| 319 |
+
},
|
| 320 |
+
{
|
| 321 |
+
"epoch": 0.3,
|
| 322 |
+
"grad_norm": 1.2366664409637451,
|
| 323 |
+
"learning_rate": 4.120528902141038e-05,
|
| 324 |
+
"loss": 1.101635456085205,
|
| 325 |
+
"step": 900
|
| 326 |
+
},
|
| 327 |
+
{
|
| 328 |
+
"epoch": 0.30666666666666664,
|
| 329 |
+
"grad_norm": 1.1046019792556763,
|
| 330 |
+
"learning_rate": 4.0789069936048135e-05,
|
| 331 |
+
"loss": 1.0482959747314453,
|
| 332 |
+
"step": 920
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.31333333333333335,
|
| 336 |
+
"grad_norm": 1.8129781484603882,
|
| 337 |
+
"learning_rate": 4.036543939868533e-05,
|
| 338 |
+
"loss": 1.2192636489868165,
|
| 339 |
+
"step": 940
|
| 340 |
+
},
|
| 341 |
+
{
|
| 342 |
+
"epoch": 0.32,
|
| 343 |
+
"grad_norm": 0.8395664691925049,
|
| 344 |
+
"learning_rate": 3.993459626317506e-05,
|
| 345 |
+
"loss": 0.8940506935119629,
|
| 346 |
+
"step": 960
|
| 347 |
+
},
|
| 348 |
+
{
|
| 349 |
+
"epoch": 0.32666666666666666,
|
| 350 |
+
"grad_norm": 1.2955873012542725,
|
| 351 |
+
"learning_rate": 3.949674276899257e-05,
|
| 352 |
+
"loss": 0.7923289775848389,
|
| 353 |
+
"step": 980
|
| 354 |
+
},
|
| 355 |
+
{
|
| 356 |
+
"epoch": 0.3333333333333333,
|
| 357 |
+
"grad_norm": 1.2329128980636597,
|
| 358 |
+
"learning_rate": 3.905208444630327e-05,
|
| 359 |
+
"loss": 0.6925106525421143,
|
| 360 |
+
"step": 1000
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 0.34,
|
| 364 |
+
"grad_norm": 1.7410433292388916,
|
| 365 |
+
"learning_rate": 3.8600830019486013e-05,
|
| 366 |
+
"loss": 1.0536805152893067,
|
| 367 |
+
"step": 1020
|
| 368 |
+
},
|
| 369 |
+
{
|
| 370 |
+
"epoch": 0.3466666666666667,
|
| 371 |
+
"grad_norm": 1.0676844120025635,
|
| 372 |
+
"learning_rate": 3.814319130915709e-05,
|
| 373 |
+
"loss": 1.2916886329650878,
|
| 374 |
+
"step": 1040
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.35333333333333333,
|
| 378 |
+
"grad_norm": 1.0247275829315186,
|
| 379 |
+
"learning_rate": 3.7679383132740767e-05,
|
| 380 |
+
"loss": 1.041970157623291,
|
| 381 |
+
"step": 1060
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 0.36,
|
| 385 |
+
"grad_norm": 1.2280709743499756,
|
| 386 |
+
"learning_rate": 3.720962320363313e-05,
|
| 387 |
+
"loss": 0.7665971755981446,
|
| 388 |
+
"step": 1080
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 0.36666666666666664,
|
| 392 |
+
"grad_norm": 1.792985200881958,
|
| 393 |
+
"learning_rate": 3.673413202900659e-05,
|
| 394 |
+
"loss": 0.9495917320251465,
|
| 395 |
+
"step": 1100
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 0.37333333333333335,
|
| 399 |
+
"grad_norm": 0.9368265867233276,
|
| 400 |
+
"learning_rate": 3.6253132806302895e-05,
|
| 401 |
+
"loss": 0.7774296283721924,
|
| 402 |
+
"step": 1120
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 0.38,
|
| 406 |
+
"grad_norm": 1.2027090787887573,
|
| 407 |
+
"learning_rate": 3.576685131846339e-05,
|
| 408 |
+
"loss": 0.9888124465942383,
|
| 409 |
+
"step": 1140
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 0.38666666666666666,
|
| 413 |
+
"grad_norm": 1.8240844011306763,
|
| 414 |
+
"learning_rate": 3.5275515827945635e-05,
|
| 415 |
+
"loss": 1.134997844696045,
|
| 416 |
+
"step": 1160
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.3933333333333333,
|
| 420 |
+
"grad_norm": 0.8984581828117371,
|
| 421 |
+
"learning_rate": 3.4779356969576044e-05,
|
| 422 |
+
"loss": 0.6457856655120849,
|
| 423 |
+
"step": 1180
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 0.4,
|
| 427 |
+
"grad_norm": 0.8814728260040283,
|
| 428 |
+
"learning_rate": 3.427860764228903e-05,
|
| 429 |
+
"loss": 0.8272159576416016,
|
| 430 |
+
"step": 1200
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 0.4066666666666667,
|
| 434 |
+
"grad_norm": 1.533231496810913,
|
| 435 |
+
"learning_rate": 3.377350289980331e-05,
|
| 436 |
+
"loss": 0.7864272117614746,
|
| 437 |
+
"step": 1220
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 0.41333333333333333,
|
| 441 |
+
"grad_norm": 1.4076637029647827,
|
| 442 |
+
"learning_rate": 3.326427984028675e-05,
|
| 443 |
+
"loss": 0.9194063186645508,
|
| 444 |
+
"step": 1240
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 0.42,
|
| 448 |
+
"grad_norm": 1.770647406578064,
|
| 449 |
+
"learning_rate": 3.275117749506156e-05,
|
| 450 |
+
"loss": 1.0757792472839356,
|
| 451 |
+
"step": 1260
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 0.4266666666666667,
|
| 455 |
+
"grad_norm": 1.488553524017334,
|
| 456 |
+
"learning_rate": 3.2234436716402025e-05,
|
| 457 |
+
"loss": 1.0499624252319335,
|
| 458 |
+
"step": 1280
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.43333333333333335,
|
| 462 |
+
"grad_norm": 0.9816551804542542,
|
| 463 |
+
"learning_rate": 3.171430006447751e-05,
|
| 464 |
+
"loss": 1.106693172454834,
|
| 465 |
+
"step": 1300
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 0.44,
|
| 469 |
+
"grad_norm": 0.9064175486564636,
|
| 470 |
+
"learning_rate": 3.119101169349374e-05,
|
| 471 |
+
"loss": 0.692702341079712,
|
| 472 |
+
"step": 1320
|
| 473 |
+
},
|
| 474 |
+
{
|
| 475 |
+
"epoch": 0.44666666666666666,
|
| 476 |
+
"grad_norm": 0.8113229870796204,
|
| 477 |
+
"learning_rate": 3.066481723708587e-05,
|
| 478 |
+
"loss": 0.41944007873535155,
|
| 479 |
+
"step": 1340
|
| 480 |
+
},
|
| 481 |
+
{
|
| 482 |
+
"epoch": 0.4533333333333333,
|
| 483 |
+
"grad_norm": 0.6801184415817261,
|
| 484 |
+
"learning_rate": 3.013596369301706e-05,
|
| 485 |
+
"loss": 0.26016736030578613,
|
| 486 |
+
"step": 1360
|
| 487 |
+
},
|
| 488 |
+
{
|
| 489 |
+
"epoch": 0.46,
|
| 490 |
+
"grad_norm": 0.9605411887168884,
|
| 491 |
+
"learning_rate": 2.9604699307236765e-05,
|
| 492 |
+
"loss": 0.47470459938049314,
|
| 493 |
+
"step": 1380
|
| 494 |
+
},
|
| 495 |
+
{
|
| 496 |
+
"epoch": 0.4666666666666667,
|
| 497 |
+
"grad_norm": 1.1666914224624634,
|
| 498 |
+
"learning_rate": 2.9071273457353082e-05,
|
| 499 |
+
"loss": 0.5180130481719971,
|
| 500 |
+
"step": 1400
|
| 501 |
+
}
|
| 502 |
+
],
|
| 503 |
+
"logging_steps": 20,
|
| 504 |
+
"max_steps": 3000,
|
| 505 |
+
"num_input_tokens_seen": 0,
|
| 506 |
+
"num_train_epochs": 9223372036854775807,
|
| 507 |
+
"save_steps": 200,
|
| 508 |
+
"stateful_callbacks": {
|
| 509 |
+
"TrainerControl": {
|
| 510 |
+
"args": {
|
| 511 |
+
"should_epoch_stop": false,
|
| 512 |
+
"should_evaluate": false,
|
| 513 |
+
"should_log": false,
|
| 514 |
+
"should_save": true,
|
| 515 |
+
"should_training_stop": false
|
| 516 |
+
},
|
| 517 |
+
"attributes": {}
|
| 518 |
+
}
|
| 519 |
+
},
|
| 520 |
+
"total_flos": 2926523606630400.0,
|
| 521 |
+
"train_batch_size": 4,
|
| 522 |
+
"trial_name": null,
|
| 523 |
+
"trial_params": null
|
| 524 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e17dad6e36ce8a3ae8352d5828d513b0bc50da179fdf9c3d376a0b3a0a20d3a5
|
| 3 |
+
size 5201
|