Upload 4 files
Browse files- config.json +38 -0
- scheduler.pt +3 -0
- trainer_state.json +718 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,38 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 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 |
+
"summary_activation": null,
|
| 25 |
+
"summary_first_dropout": 0.1,
|
| 26 |
+
"summary_proj_to_labels": true,
|
| 27 |
+
"summary_type": "cls_index",
|
| 28 |
+
"summary_use_proj": true,
|
| 29 |
+
"task_specific_params": {
|
| 30 |
+
"text-generation": {
|
| 31 |
+
"do_sample": true,
|
| 32 |
+
"max_length": 50
|
| 33 |
+
}
|
| 34 |
+
},
|
| 35 |
+
"transformers_version": "4.4.2",
|
| 36 |
+
"use_cache": true,
|
| 37 |
+
"vocab_size": 50259
|
| 38 |
+
}
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d4a097880e6ec24e08b3d9a61c6cdcc3f693e6e713c9aba7b89fa76154ac28e8
|
| 3 |
+
size 623
|
trainer_state.json
ADDED
|
@@ -0,0 +1,718 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.9893622418779279,
|
| 5 |
+
"global_step": 58500,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.01,
|
| 12 |
+
"learning_rate": 4.9585581623505137e-05,
|
| 13 |
+
"loss": 0.6366,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.02,
|
| 18 |
+
"learning_rate": 4.9162705729122623e-05,
|
| 19 |
+
"loss": 0.2794,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.03,
|
| 24 |
+
"learning_rate": 4.8739829834740104e-05,
|
| 25 |
+
"loss": 0.2335,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.03,
|
| 30 |
+
"learning_rate": 4.8316953940357584e-05,
|
| 31 |
+
"loss": 0.2659,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.04,
|
| 36 |
+
"learning_rate": 4.789407804597507e-05,
|
| 37 |
+
"loss": 0.2433,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.05,
|
| 42 |
+
"learning_rate": 4.747120215159255e-05,
|
| 43 |
+
"loss": 0.2383,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.06,
|
| 48 |
+
"learning_rate": 4.704832625721004e-05,
|
| 49 |
+
"loss": 0.2382,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.07,
|
| 54 |
+
"learning_rate": 4.6625450362827524e-05,
|
| 55 |
+
"loss": 0.2224,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.08,
|
| 60 |
+
"learning_rate": 4.6202574468445005e-05,
|
| 61 |
+
"loss": 0.2412,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.08,
|
| 66 |
+
"learning_rate": 4.5779698574062485e-05,
|
| 67 |
+
"loss": 0.2221,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.09,
|
| 72 |
+
"learning_rate": 4.535682267967997e-05,
|
| 73 |
+
"loss": 0.2365,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.1,
|
| 78 |
+
"learning_rate": 4.493394678529745e-05,
|
| 79 |
+
"loss": 0.206,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.11,
|
| 84 |
+
"learning_rate": 4.451107089091494e-05,
|
| 85 |
+
"loss": 0.2189,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.12,
|
| 90 |
+
"learning_rate": 4.408819499653242e-05,
|
| 91 |
+
"loss": 0.2206,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.13,
|
| 96 |
+
"learning_rate": 4.3665319102149906e-05,
|
| 97 |
+
"loss": 0.2238,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.14,
|
| 102 |
+
"learning_rate": 4.3242443207767386e-05,
|
| 103 |
+
"loss": 0.2127,
|
| 104 |
+
"step": 8000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.14,
|
| 108 |
+
"learning_rate": 4.2819567313384866e-05,
|
| 109 |
+
"loss": 0.2103,
|
| 110 |
+
"step": 8500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.15,
|
| 114 |
+
"learning_rate": 4.239669141900235e-05,
|
| 115 |
+
"loss": 0.216,
|
| 116 |
+
"step": 9000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.16,
|
| 120 |
+
"learning_rate": 4.197381552461983e-05,
|
| 121 |
+
"loss": 0.1916,
|
| 122 |
+
"step": 9500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.17,
|
| 126 |
+
"learning_rate": 4.155093963023732e-05,
|
| 127 |
+
"loss": 0.2208,
|
| 128 |
+
"step": 10000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.18,
|
| 132 |
+
"learning_rate": 4.1128063735854807e-05,
|
| 133 |
+
"loss": 0.1848,
|
| 134 |
+
"step": 10500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.19,
|
| 138 |
+
"learning_rate": 4.070518784147229e-05,
|
| 139 |
+
"loss": 0.2126,
|
| 140 |
+
"step": 11000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.19,
|
| 144 |
+
"learning_rate": 4.028231194708977e-05,
|
| 145 |
+
"loss": 0.1963,
|
| 146 |
+
"step": 11500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.2,
|
| 150 |
+
"learning_rate": 3.9859436052707254e-05,
|
| 151 |
+
"loss": 0.2159,
|
| 152 |
+
"step": 12000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.21,
|
| 156 |
+
"learning_rate": 3.9436560158324734e-05,
|
| 157 |
+
"loss": 0.1872,
|
| 158 |
+
"step": 12500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 0.22,
|
| 162 |
+
"learning_rate": 3.901368426394222e-05,
|
| 163 |
+
"loss": 0.2261,
|
| 164 |
+
"step": 13000
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.23,
|
| 168 |
+
"learning_rate": 3.859080836955971e-05,
|
| 169 |
+
"loss": 0.2147,
|
| 170 |
+
"step": 13500
|
| 171 |
+
},
|
| 172 |
+
{
|
| 173 |
+
"epoch": 0.24,
|
| 174 |
+
"learning_rate": 3.816793247517719e-05,
|
| 175 |
+
"loss": 0.1938,
|
| 176 |
+
"step": 14000
|
| 177 |
+
},
|
| 178 |
+
{
|
| 179 |
+
"epoch": 0.25,
|
| 180 |
+
"learning_rate": 3.774505658079467e-05,
|
| 181 |
+
"loss": 0.1938,
|
| 182 |
+
"step": 14500
|
| 183 |
+
},
|
| 184 |
+
{
|
| 185 |
+
"epoch": 0.25,
|
| 186 |
+
"learning_rate": 3.7322180686412155e-05,
|
| 187 |
+
"loss": 0.1935,
|
| 188 |
+
"step": 15000
|
| 189 |
+
},
|
| 190 |
+
{
|
| 191 |
+
"epoch": 0.26,
|
| 192 |
+
"learning_rate": 3.6899304792029635e-05,
|
| 193 |
+
"loss": 0.209,
|
| 194 |
+
"step": 15500
|
| 195 |
+
},
|
| 196 |
+
{
|
| 197 |
+
"epoch": 0.27,
|
| 198 |
+
"learning_rate": 3.647642889764712e-05,
|
| 199 |
+
"loss": 0.1981,
|
| 200 |
+
"step": 16000
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.28,
|
| 204 |
+
"learning_rate": 3.605355300326461e-05,
|
| 205 |
+
"loss": 0.2031,
|
| 206 |
+
"step": 16500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.29,
|
| 210 |
+
"learning_rate": 3.563067710888209e-05,
|
| 211 |
+
"loss": 0.2045,
|
| 212 |
+
"step": 17000
|
| 213 |
+
},
|
| 214 |
+
{
|
| 215 |
+
"epoch": 0.3,
|
| 216 |
+
"learning_rate": 3.520780121449957e-05,
|
| 217 |
+
"loss": 0.1931,
|
| 218 |
+
"step": 17500
|
| 219 |
+
},
|
| 220 |
+
{
|
| 221 |
+
"epoch": 0.3,
|
| 222 |
+
"learning_rate": 3.478492532011705e-05,
|
| 223 |
+
"loss": 0.1936,
|
| 224 |
+
"step": 18000
|
| 225 |
+
},
|
| 226 |
+
{
|
| 227 |
+
"epoch": 0.31,
|
| 228 |
+
"learning_rate": 3.4362049425734536e-05,
|
| 229 |
+
"loss": 0.2128,
|
| 230 |
+
"step": 18500
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 0.32,
|
| 234 |
+
"learning_rate": 3.393917353135202e-05,
|
| 235 |
+
"loss": 0.1841,
|
| 236 |
+
"step": 19000
|
| 237 |
+
},
|
| 238 |
+
{
|
| 239 |
+
"epoch": 0.33,
|
| 240 |
+
"learning_rate": 3.35162976369695e-05,
|
| 241 |
+
"loss": 0.193,
|
| 242 |
+
"step": 19500
|
| 243 |
+
},
|
| 244 |
+
{
|
| 245 |
+
"epoch": 0.34,
|
| 246 |
+
"learning_rate": 3.309342174258699e-05,
|
| 247 |
+
"loss": 0.1961,
|
| 248 |
+
"step": 20000
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.35,
|
| 252 |
+
"learning_rate": 3.267054584820447e-05,
|
| 253 |
+
"loss": 0.1913,
|
| 254 |
+
"step": 20500
|
| 255 |
+
},
|
| 256 |
+
{
|
| 257 |
+
"epoch": 0.36,
|
| 258 |
+
"learning_rate": 3.224766995382195e-05,
|
| 259 |
+
"loss": 0.2081,
|
| 260 |
+
"step": 21000
|
| 261 |
+
},
|
| 262 |
+
{
|
| 263 |
+
"epoch": 0.36,
|
| 264 |
+
"learning_rate": 3.182479405943944e-05,
|
| 265 |
+
"loss": 0.1866,
|
| 266 |
+
"step": 21500
|
| 267 |
+
},
|
| 268 |
+
{
|
| 269 |
+
"epoch": 0.37,
|
| 270 |
+
"learning_rate": 3.1401918165056924e-05,
|
| 271 |
+
"loss": 0.1884,
|
| 272 |
+
"step": 22000
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 0.38,
|
| 276 |
+
"learning_rate": 3.0979042270674404e-05,
|
| 277 |
+
"loss": 0.1936,
|
| 278 |
+
"step": 22500
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.39,
|
| 282 |
+
"learning_rate": 3.055616637629189e-05,
|
| 283 |
+
"loss": 0.2044,
|
| 284 |
+
"step": 23000
|
| 285 |
+
},
|
| 286 |
+
{
|
| 287 |
+
"epoch": 0.4,
|
| 288 |
+
"learning_rate": 3.013329048190937e-05,
|
| 289 |
+
"loss": 0.1916,
|
| 290 |
+
"step": 23500
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.41,
|
| 294 |
+
"learning_rate": 2.971041458752685e-05,
|
| 295 |
+
"loss": 0.1787,
|
| 296 |
+
"step": 24000
|
| 297 |
+
},
|
| 298 |
+
{
|
| 299 |
+
"epoch": 0.41,
|
| 300 |
+
"learning_rate": 2.928753869314434e-05,
|
| 301 |
+
"loss": 0.2117,
|
| 302 |
+
"step": 24500
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 0.42,
|
| 306 |
+
"learning_rate": 2.886466279876182e-05,
|
| 307 |
+
"loss": 0.1983,
|
| 308 |
+
"step": 25000
|
| 309 |
+
},
|
| 310 |
+
{
|
| 311 |
+
"epoch": 0.43,
|
| 312 |
+
"learning_rate": 2.84417869043793e-05,
|
| 313 |
+
"loss": 0.1947,
|
| 314 |
+
"step": 25500
|
| 315 |
+
},
|
| 316 |
+
{
|
| 317 |
+
"epoch": 0.44,
|
| 318 |
+
"learning_rate": 2.801891100999679e-05,
|
| 319 |
+
"loss": 0.1919,
|
| 320 |
+
"step": 26000
|
| 321 |
+
},
|
| 322 |
+
{
|
| 323 |
+
"epoch": 0.45,
|
| 324 |
+
"learning_rate": 2.7596035115614272e-05,
|
| 325 |
+
"loss": 0.1799,
|
| 326 |
+
"step": 26500
|
| 327 |
+
},
|
| 328 |
+
{
|
| 329 |
+
"epoch": 0.46,
|
| 330 |
+
"learning_rate": 2.7173159221231752e-05,
|
| 331 |
+
"loss": 0.1746,
|
| 332 |
+
"step": 27000
|
| 333 |
+
},
|
| 334 |
+
{
|
| 335 |
+
"epoch": 0.47,
|
| 336 |
+
"learning_rate": 2.675028332684924e-05,
|
| 337 |
+
"loss": 0.1768,
|
| 338 |
+
"step": 27500
|
| 339 |
+
},
|
| 340 |
+
{
|
| 341 |
+
"epoch": 0.47,
|
| 342 |
+
"learning_rate": 2.6327407432466722e-05,
|
| 343 |
+
"loss": 0.1758,
|
| 344 |
+
"step": 28000
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 0.48,
|
| 348 |
+
"learning_rate": 2.5904531538084202e-05,
|
| 349 |
+
"loss": 0.1882,
|
| 350 |
+
"step": 28500
|
| 351 |
+
},
|
| 352 |
+
{
|
| 353 |
+
"epoch": 0.49,
|
| 354 |
+
"learning_rate": 2.5481655643701686e-05,
|
| 355 |
+
"loss": 0.1816,
|
| 356 |
+
"step": 29000
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.5,
|
| 360 |
+
"learning_rate": 2.5058779749319173e-05,
|
| 361 |
+
"loss": 0.197,
|
| 362 |
+
"step": 29500
|
| 363 |
+
},
|
| 364 |
+
{
|
| 365 |
+
"epoch": 0.51,
|
| 366 |
+
"learning_rate": 2.4635903854936653e-05,
|
| 367 |
+
"loss": 0.1672,
|
| 368 |
+
"step": 30000
|
| 369 |
+
},
|
| 370 |
+
{
|
| 371 |
+
"epoch": 0.52,
|
| 372 |
+
"learning_rate": 2.421302796055414e-05,
|
| 373 |
+
"loss": 0.1709,
|
| 374 |
+
"step": 30500
|
| 375 |
+
},
|
| 376 |
+
{
|
| 377 |
+
"epoch": 0.52,
|
| 378 |
+
"learning_rate": 2.379015206617162e-05,
|
| 379 |
+
"loss": 0.1688,
|
| 380 |
+
"step": 31000
|
| 381 |
+
},
|
| 382 |
+
{
|
| 383 |
+
"epoch": 0.53,
|
| 384 |
+
"learning_rate": 2.3367276171789103e-05,
|
| 385 |
+
"loss": 0.1945,
|
| 386 |
+
"step": 31500
|
| 387 |
+
},
|
| 388 |
+
{
|
| 389 |
+
"epoch": 0.54,
|
| 390 |
+
"learning_rate": 2.2944400277406587e-05,
|
| 391 |
+
"loss": 0.1888,
|
| 392 |
+
"step": 32000
|
| 393 |
+
},
|
| 394 |
+
{
|
| 395 |
+
"epoch": 0.55,
|
| 396 |
+
"learning_rate": 2.252152438302407e-05,
|
| 397 |
+
"loss": 0.1634,
|
| 398 |
+
"step": 32500
|
| 399 |
+
},
|
| 400 |
+
{
|
| 401 |
+
"epoch": 0.56,
|
| 402 |
+
"learning_rate": 2.2098648488641554e-05,
|
| 403 |
+
"loss": 0.2004,
|
| 404 |
+
"step": 33000
|
| 405 |
+
},
|
| 406 |
+
{
|
| 407 |
+
"epoch": 0.57,
|
| 408 |
+
"learning_rate": 2.1675772594259037e-05,
|
| 409 |
+
"loss": 0.1886,
|
| 410 |
+
"step": 33500
|
| 411 |
+
},
|
| 412 |
+
{
|
| 413 |
+
"epoch": 0.58,
|
| 414 |
+
"learning_rate": 2.125289669987652e-05,
|
| 415 |
+
"loss": 0.1814,
|
| 416 |
+
"step": 34000
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 0.58,
|
| 420 |
+
"learning_rate": 2.0830020805494004e-05,
|
| 421 |
+
"loss": 0.1939,
|
| 422 |
+
"step": 34500
|
| 423 |
+
},
|
| 424 |
+
{
|
| 425 |
+
"epoch": 0.59,
|
| 426 |
+
"learning_rate": 2.0407144911111488e-05,
|
| 427 |
+
"loss": 0.1932,
|
| 428 |
+
"step": 35000
|
| 429 |
+
},
|
| 430 |
+
{
|
| 431 |
+
"epoch": 0.6,
|
| 432 |
+
"learning_rate": 1.998426901672897e-05,
|
| 433 |
+
"loss": 0.1694,
|
| 434 |
+
"step": 35500
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.61,
|
| 438 |
+
"learning_rate": 1.9561393122346455e-05,
|
| 439 |
+
"loss": 0.1883,
|
| 440 |
+
"step": 36000
|
| 441 |
+
},
|
| 442 |
+
{
|
| 443 |
+
"epoch": 0.62,
|
| 444 |
+
"learning_rate": 1.913851722796394e-05,
|
| 445 |
+
"loss": 0.1623,
|
| 446 |
+
"step": 36500
|
| 447 |
+
},
|
| 448 |
+
{
|
| 449 |
+
"epoch": 0.63,
|
| 450 |
+
"learning_rate": 1.8715641333581422e-05,
|
| 451 |
+
"loss": 0.1703,
|
| 452 |
+
"step": 37000
|
| 453 |
+
},
|
| 454 |
+
{
|
| 455 |
+
"epoch": 0.63,
|
| 456 |
+
"learning_rate": 1.8292765439198905e-05,
|
| 457 |
+
"loss": 0.1795,
|
| 458 |
+
"step": 37500
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 0.64,
|
| 462 |
+
"learning_rate": 1.7869889544816386e-05,
|
| 463 |
+
"loss": 0.1873,
|
| 464 |
+
"step": 38000
|
| 465 |
+
},
|
| 466 |
+
{
|
| 467 |
+
"epoch": 0.65,
|
| 468 |
+
"learning_rate": 1.7447013650433872e-05,
|
| 469 |
+
"loss": 0.1657,
|
| 470 |
+
"step": 38500
|
| 471 |
+
},
|
| 472 |
+
{
|
| 473 |
+
"epoch": 0.66,
|
| 474 |
+
"learning_rate": 1.7024137756051356e-05,
|
| 475 |
+
"loss": 0.1806,
|
| 476 |
+
"step": 39000
|
| 477 |
+
},
|
| 478 |
+
{
|
| 479 |
+
"epoch": 0.67,
|
| 480 |
+
"learning_rate": 1.6601261861668836e-05,
|
| 481 |
+
"loss": 0.1701,
|
| 482 |
+
"step": 39500
|
| 483 |
+
},
|
| 484 |
+
{
|
| 485 |
+
"epoch": 0.68,
|
| 486 |
+
"learning_rate": 1.6178385967286323e-05,
|
| 487 |
+
"loss": 0.1646,
|
| 488 |
+
"step": 40000
|
| 489 |
+
},
|
| 490 |
+
{
|
| 491 |
+
"epoch": 0.68,
|
| 492 |
+
"learning_rate": 1.5755510072903806e-05,
|
| 493 |
+
"loss": 0.1686,
|
| 494 |
+
"step": 40500
|
| 495 |
+
},
|
| 496 |
+
{
|
| 497 |
+
"epoch": 0.69,
|
| 498 |
+
"learning_rate": 1.5332634178521287e-05,
|
| 499 |
+
"loss": 0.1752,
|
| 500 |
+
"step": 41000
|
| 501 |
+
},
|
| 502 |
+
{
|
| 503 |
+
"epoch": 0.7,
|
| 504 |
+
"learning_rate": 1.4909758284138772e-05,
|
| 505 |
+
"loss": 0.1631,
|
| 506 |
+
"step": 41500
|
| 507 |
+
},
|
| 508 |
+
{
|
| 509 |
+
"epoch": 0.71,
|
| 510 |
+
"learning_rate": 1.4486882389756254e-05,
|
| 511 |
+
"loss": 0.1645,
|
| 512 |
+
"step": 42000
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.72,
|
| 516 |
+
"learning_rate": 1.4064006495373739e-05,
|
| 517 |
+
"loss": 0.1815,
|
| 518 |
+
"step": 42500
|
| 519 |
+
},
|
| 520 |
+
{
|
| 521 |
+
"epoch": 0.73,
|
| 522 |
+
"learning_rate": 1.3641130600991222e-05,
|
| 523 |
+
"loss": 0.1928,
|
| 524 |
+
"step": 43000
|
| 525 |
+
},
|
| 526 |
+
{
|
| 527 |
+
"epoch": 0.74,
|
| 528 |
+
"learning_rate": 1.3218254706608704e-05,
|
| 529 |
+
"loss": 0.1637,
|
| 530 |
+
"step": 43500
|
| 531 |
+
},
|
| 532 |
+
{
|
| 533 |
+
"epoch": 0.74,
|
| 534 |
+
"learning_rate": 1.279537881222619e-05,
|
| 535 |
+
"loss": 0.1658,
|
| 536 |
+
"step": 44000
|
| 537 |
+
},
|
| 538 |
+
{
|
| 539 |
+
"epoch": 0.75,
|
| 540 |
+
"learning_rate": 1.2372502917843671e-05,
|
| 541 |
+
"loss": 0.1754,
|
| 542 |
+
"step": 44500
|
| 543 |
+
},
|
| 544 |
+
{
|
| 545 |
+
"epoch": 0.76,
|
| 546 |
+
"learning_rate": 1.1949627023461155e-05,
|
| 547 |
+
"loss": 0.1787,
|
| 548 |
+
"step": 45000
|
| 549 |
+
},
|
| 550 |
+
{
|
| 551 |
+
"epoch": 0.77,
|
| 552 |
+
"learning_rate": 1.152675112907864e-05,
|
| 553 |
+
"loss": 0.1499,
|
| 554 |
+
"step": 45500
|
| 555 |
+
},
|
| 556 |
+
{
|
| 557 |
+
"epoch": 0.78,
|
| 558 |
+
"learning_rate": 1.1103875234696122e-05,
|
| 559 |
+
"loss": 0.1659,
|
| 560 |
+
"step": 46000
|
| 561 |
+
},
|
| 562 |
+
{
|
| 563 |
+
"epoch": 0.79,
|
| 564 |
+
"learning_rate": 1.0680999340313605e-05,
|
| 565 |
+
"loss": 0.175,
|
| 566 |
+
"step": 46500
|
| 567 |
+
},
|
| 568 |
+
{
|
| 569 |
+
"epoch": 0.79,
|
| 570 |
+
"learning_rate": 1.0258123445931089e-05,
|
| 571 |
+
"loss": 0.1659,
|
| 572 |
+
"step": 47000
|
| 573 |
+
},
|
| 574 |
+
{
|
| 575 |
+
"epoch": 0.8,
|
| 576 |
+
"learning_rate": 9.835247551548572e-06,
|
| 577 |
+
"loss": 0.1904,
|
| 578 |
+
"step": 47500
|
| 579 |
+
},
|
| 580 |
+
{
|
| 581 |
+
"epoch": 0.81,
|
| 582 |
+
"learning_rate": 9.412371657166056e-06,
|
| 583 |
+
"loss": 0.1736,
|
| 584 |
+
"step": 48000
|
| 585 |
+
},
|
| 586 |
+
{
|
| 587 |
+
"epoch": 0.82,
|
| 588 |
+
"learning_rate": 8.989495762783539e-06,
|
| 589 |
+
"loss": 0.1873,
|
| 590 |
+
"step": 48500
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.83,
|
| 594 |
+
"learning_rate": 8.566619868401023e-06,
|
| 595 |
+
"loss": 0.1809,
|
| 596 |
+
"step": 49000
|
| 597 |
+
},
|
| 598 |
+
{
|
| 599 |
+
"epoch": 0.84,
|
| 600 |
+
"learning_rate": 8.143743974018506e-06,
|
| 601 |
+
"loss": 0.1714,
|
| 602 |
+
"step": 49500
|
| 603 |
+
},
|
| 604 |
+
{
|
| 605 |
+
"epoch": 0.85,
|
| 606 |
+
"learning_rate": 7.720868079635988e-06,
|
| 607 |
+
"loss": 0.162,
|
| 608 |
+
"step": 50000
|
| 609 |
+
},
|
| 610 |
+
{
|
| 611 |
+
"epoch": 0.85,
|
| 612 |
+
"learning_rate": 7.297992185253471e-06,
|
| 613 |
+
"loss": 0.1758,
|
| 614 |
+
"step": 50500
|
| 615 |
+
},
|
| 616 |
+
{
|
| 617 |
+
"epoch": 0.86,
|
| 618 |
+
"learning_rate": 6.875116290870956e-06,
|
| 619 |
+
"loss": 0.1705,
|
| 620 |
+
"step": 51000
|
| 621 |
+
},
|
| 622 |
+
{
|
| 623 |
+
"epoch": 0.87,
|
| 624 |
+
"learning_rate": 6.452240396488439e-06,
|
| 625 |
+
"loss": 0.1632,
|
| 626 |
+
"step": 51500
|
| 627 |
+
},
|
| 628 |
+
{
|
| 629 |
+
"epoch": 0.88,
|
| 630 |
+
"learning_rate": 6.029364502105922e-06,
|
| 631 |
+
"loss": 0.1767,
|
| 632 |
+
"step": 52000
|
| 633 |
+
},
|
| 634 |
+
{
|
| 635 |
+
"epoch": 0.89,
|
| 636 |
+
"learning_rate": 5.606488607723405e-06,
|
| 637 |
+
"loss": 0.1705,
|
| 638 |
+
"step": 52500
|
| 639 |
+
},
|
| 640 |
+
{
|
| 641 |
+
"epoch": 0.9,
|
| 642 |
+
"learning_rate": 5.183612713340889e-06,
|
| 643 |
+
"loss": 0.1636,
|
| 644 |
+
"step": 53000
|
| 645 |
+
},
|
| 646 |
+
{
|
| 647 |
+
"epoch": 0.9,
|
| 648 |
+
"learning_rate": 4.760736818958372e-06,
|
| 649 |
+
"loss": 0.1666,
|
| 650 |
+
"step": 53500
|
| 651 |
+
},
|
| 652 |
+
{
|
| 653 |
+
"epoch": 0.91,
|
| 654 |
+
"learning_rate": 4.337860924575856e-06,
|
| 655 |
+
"loss": 0.1765,
|
| 656 |
+
"step": 54000
|
| 657 |
+
},
|
| 658 |
+
{
|
| 659 |
+
"epoch": 0.92,
|
| 660 |
+
"learning_rate": 3.9149850301933386e-06,
|
| 661 |
+
"loss": 0.1634,
|
| 662 |
+
"step": 54500
|
| 663 |
+
},
|
| 664 |
+
{
|
| 665 |
+
"epoch": 0.93,
|
| 666 |
+
"learning_rate": 3.4921091358108225e-06,
|
| 667 |
+
"loss": 0.159,
|
| 668 |
+
"step": 55000
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.94,
|
| 672 |
+
"learning_rate": 3.0692332414283056e-06,
|
| 673 |
+
"loss": 0.1725,
|
| 674 |
+
"step": 55500
|
| 675 |
+
},
|
| 676 |
+
{
|
| 677 |
+
"epoch": 0.95,
|
| 678 |
+
"learning_rate": 2.646357347045789e-06,
|
| 679 |
+
"loss": 0.1568,
|
| 680 |
+
"step": 56000
|
| 681 |
+
},
|
| 682 |
+
{
|
| 683 |
+
"epoch": 0.96,
|
| 684 |
+
"learning_rate": 2.223481452663272e-06,
|
| 685 |
+
"loss": 0.165,
|
| 686 |
+
"step": 56500
|
| 687 |
+
},
|
| 688 |
+
{
|
| 689 |
+
"epoch": 0.96,
|
| 690 |
+
"learning_rate": 1.800605558280756e-06,
|
| 691 |
+
"loss": 0.1825,
|
| 692 |
+
"step": 57000
|
| 693 |
+
},
|
| 694 |
+
{
|
| 695 |
+
"epoch": 0.97,
|
| 696 |
+
"learning_rate": 1.3777296638982391e-06,
|
| 697 |
+
"loss": 0.1646,
|
| 698 |
+
"step": 57500
|
| 699 |
+
},
|
| 700 |
+
{
|
| 701 |
+
"epoch": 0.98,
|
| 702 |
+
"learning_rate": 9.548537695157224e-07,
|
| 703 |
+
"loss": 0.173,
|
| 704 |
+
"step": 58000
|
| 705 |
+
},
|
| 706 |
+
{
|
| 707 |
+
"epoch": 0.99,
|
| 708 |
+
"learning_rate": 5.319778751332059e-07,
|
| 709 |
+
"loss": 0.1723,
|
| 710 |
+
"step": 58500
|
| 711 |
+
}
|
| 712 |
+
],
|
| 713 |
+
"max_steps": 59129,
|
| 714 |
+
"num_train_epochs": 1,
|
| 715 |
+
"total_flos": 1.27532698435584e+17,
|
| 716 |
+
"trial_name": null,
|
| 717 |
+
"trial_params": null
|
| 718 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:22faf7ed1d3602868e0f6fc66eddbb4b9086f0909af678c53793d8f24b0b19be
|
| 3 |
+
size 2223
|