Commit ·
4f6ce4b
1
Parent(s): 19b86a3
Upload V5 of the model
Browse filesThis is our upload of the EndoGPT model.
- config.json +27 -0
- generation_config.json +7 -0
- optimizer.pt +3 -0
- pytorch_model.bin +3 -0
- rng_state.pth +3 -0
- scheduler.pt +3 -0
- trainer_state.json +488 -0
- training_args.bin +3 -0
config.json
ADDED
|
@@ -0,0 +1,27 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "models/checkpoint-38000/",
|
| 3 |
+
"activation_dropout": 0.0,
|
| 4 |
+
"architectures": [
|
| 5 |
+
"BioGptForCausalLM"
|
| 6 |
+
],
|
| 7 |
+
"attention_probs_dropout_prob": 0.1,
|
| 8 |
+
"bos_token_id": 0,
|
| 9 |
+
"eos_token_id": 2,
|
| 10 |
+
"hidden_act": "gelu",
|
| 11 |
+
"hidden_dropout_prob": 0.1,
|
| 12 |
+
"hidden_size": 1024,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 4096,
|
| 15 |
+
"layer_norm_eps": 1e-12,
|
| 16 |
+
"layerdrop": 0.0,
|
| 17 |
+
"max_position_embeddings": 1024,
|
| 18 |
+
"model_type": "biogpt",
|
| 19 |
+
"num_attention_heads": 16,
|
| 20 |
+
"num_hidden_layers": 24,
|
| 21 |
+
"pad_token_id": 1,
|
| 22 |
+
"scale_embedding": true,
|
| 23 |
+
"torch_dtype": "float32",
|
| 24 |
+
"transformers_version": "4.26.1",
|
| 25 |
+
"use_cache": true,
|
| 26 |
+
"vocab_size": 42384
|
| 27 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,7 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"eos_token_id": 2,
|
| 5 |
+
"pad_token_id": 1,
|
| 6 |
+
"transformers_version": "4.26.1"
|
| 7 |
+
}
|
optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:be67d508756cfe1610993380528afcdecbdf246c76e814f58f128bd473792731
|
| 3 |
+
size 2774334085
|
pytorch_model.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b83ee68b55b7d6d31f4996353dcb534282c4814d3d8cd99e1a61a310088b4531
|
| 3 |
+
size 1387181917
|
rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:63a4f76c5099492c84076495c0ea07f8fce9add91872fdf894fbf9867c660840
|
| 3 |
+
size 14639
|
scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9b0e2280cc748a4ba508dc0a51cd9af4396af6537730c71541b129b1005762b5
|
| 3 |
+
size 627
|
trainer_state.json
ADDED
|
@@ -0,0 +1,488 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 2.9625009745068995,
|
| 5 |
+
"global_step": 38000,
|
| 6 |
+
"is_hyper_param_search": false,
|
| 7 |
+
"is_local_process_zero": true,
|
| 8 |
+
"is_world_process_zero": true,
|
| 9 |
+
"log_history": [
|
| 10 |
+
{
|
| 11 |
+
"epoch": 0.04,
|
| 12 |
+
"learning_rate": 1.974013149346431e-05,
|
| 13 |
+
"loss": 0.3252,
|
| 14 |
+
"step": 500
|
| 15 |
+
},
|
| 16 |
+
{
|
| 17 |
+
"epoch": 0.08,
|
| 18 |
+
"learning_rate": 1.9480262986928614e-05,
|
| 19 |
+
"loss": 0.3841,
|
| 20 |
+
"step": 1000
|
| 21 |
+
},
|
| 22 |
+
{
|
| 23 |
+
"epoch": 0.12,
|
| 24 |
+
"learning_rate": 1.9220394480392923e-05,
|
| 25 |
+
"loss": 0.6714,
|
| 26 |
+
"step": 1500
|
| 27 |
+
},
|
| 28 |
+
{
|
| 29 |
+
"epoch": 0.16,
|
| 30 |
+
"learning_rate": 1.896052597385723e-05,
|
| 31 |
+
"loss": 0.7434,
|
| 32 |
+
"step": 2000
|
| 33 |
+
},
|
| 34 |
+
{
|
| 35 |
+
"epoch": 0.19,
|
| 36 |
+
"learning_rate": 1.8700657467321536e-05,
|
| 37 |
+
"loss": 0.3724,
|
| 38 |
+
"step": 2500
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.23,
|
| 42 |
+
"learning_rate": 1.8440788960785845e-05,
|
| 43 |
+
"loss": 0.4206,
|
| 44 |
+
"step": 3000
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.27,
|
| 48 |
+
"learning_rate": 1.818092045425015e-05,
|
| 49 |
+
"loss": 0.4183,
|
| 50 |
+
"step": 3500
|
| 51 |
+
},
|
| 52 |
+
{
|
| 53 |
+
"epoch": 0.31,
|
| 54 |
+
"learning_rate": 1.7921051947714458e-05,
|
| 55 |
+
"loss": 0.3506,
|
| 56 |
+
"step": 4000
|
| 57 |
+
},
|
| 58 |
+
{
|
| 59 |
+
"epoch": 0.35,
|
| 60 |
+
"learning_rate": 1.7661183441178764e-05,
|
| 61 |
+
"loss": 0.3634,
|
| 62 |
+
"step": 4500
|
| 63 |
+
},
|
| 64 |
+
{
|
| 65 |
+
"epoch": 0.39,
|
| 66 |
+
"learning_rate": 1.7401314934643073e-05,
|
| 67 |
+
"loss": 0.3824,
|
| 68 |
+
"step": 5000
|
| 69 |
+
},
|
| 70 |
+
{
|
| 71 |
+
"epoch": 0.43,
|
| 72 |
+
"learning_rate": 1.714144642810738e-05,
|
| 73 |
+
"loss": 0.4187,
|
| 74 |
+
"step": 5500
|
| 75 |
+
},
|
| 76 |
+
{
|
| 77 |
+
"epoch": 0.47,
|
| 78 |
+
"learning_rate": 1.6881577921571686e-05,
|
| 79 |
+
"loss": 0.3559,
|
| 80 |
+
"step": 6000
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.51,
|
| 84 |
+
"learning_rate": 1.6621709415035995e-05,
|
| 85 |
+
"loss": 0.3591,
|
| 86 |
+
"step": 6500
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 0.55,
|
| 90 |
+
"learning_rate": 1.6361840908500298e-05,
|
| 91 |
+
"loss": 0.3767,
|
| 92 |
+
"step": 7000
|
| 93 |
+
},
|
| 94 |
+
{
|
| 95 |
+
"epoch": 0.58,
|
| 96 |
+
"learning_rate": 1.6101972401964607e-05,
|
| 97 |
+
"loss": 0.3363,
|
| 98 |
+
"step": 7500
|
| 99 |
+
},
|
| 100 |
+
{
|
| 101 |
+
"epoch": 0.62,
|
| 102 |
+
"learning_rate": 1.5842103895428917e-05,
|
| 103 |
+
"loss": 0.3224,
|
| 104 |
+
"step": 8000
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.66,
|
| 108 |
+
"learning_rate": 1.558223538889322e-05,
|
| 109 |
+
"loss": 0.3362,
|
| 110 |
+
"step": 8500
|
| 111 |
+
},
|
| 112 |
+
{
|
| 113 |
+
"epoch": 0.7,
|
| 114 |
+
"learning_rate": 1.532236688235753e-05,
|
| 115 |
+
"loss": 0.3404,
|
| 116 |
+
"step": 9000
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 0.74,
|
| 120 |
+
"learning_rate": 1.5062498375821835e-05,
|
| 121 |
+
"loss": 0.3802,
|
| 122 |
+
"step": 9500
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.78,
|
| 126 |
+
"learning_rate": 1.4802629869286143e-05,
|
| 127 |
+
"loss": 0.3875,
|
| 128 |
+
"step": 10000
|
| 129 |
+
},
|
| 130 |
+
{
|
| 131 |
+
"epoch": 0.82,
|
| 132 |
+
"learning_rate": 1.4542761362750451e-05,
|
| 133 |
+
"loss": 0.4296,
|
| 134 |
+
"step": 10500
|
| 135 |
+
},
|
| 136 |
+
{
|
| 137 |
+
"epoch": 0.86,
|
| 138 |
+
"learning_rate": 1.4282892856214755e-05,
|
| 139 |
+
"loss": 0.3974,
|
| 140 |
+
"step": 11000
|
| 141 |
+
},
|
| 142 |
+
{
|
| 143 |
+
"epoch": 0.9,
|
| 144 |
+
"learning_rate": 1.4023024349679063e-05,
|
| 145 |
+
"loss": 0.3793,
|
| 146 |
+
"step": 11500
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.94,
|
| 150 |
+
"learning_rate": 1.376315584314337e-05,
|
| 151 |
+
"loss": 0.3472,
|
| 152 |
+
"step": 12000
|
| 153 |
+
},
|
| 154 |
+
{
|
| 155 |
+
"epoch": 0.97,
|
| 156 |
+
"learning_rate": 1.3503287336607677e-05,
|
| 157 |
+
"loss": 0.3598,
|
| 158 |
+
"step": 12500
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.0,
|
| 162 |
+
"eval_loss": 0.3997747600078583,
|
| 163 |
+
"eval_runtime": 662.7743,
|
| 164 |
+
"eval_samples_per_second": 66.354,
|
| 165 |
+
"eval_steps_per_second": 8.295,
|
| 166 |
+
"step": 12827
|
| 167 |
+
},
|
| 168 |
+
{
|
| 169 |
+
"epoch": 1.01,
|
| 170 |
+
"learning_rate": 1.3243418830071985e-05,
|
| 171 |
+
"loss": 0.3853,
|
| 172 |
+
"step": 13000
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 1.05,
|
| 176 |
+
"learning_rate": 1.2983550323536291e-05,
|
| 177 |
+
"loss": 0.3794,
|
| 178 |
+
"step": 13500
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 1.09,
|
| 182 |
+
"learning_rate": 1.2723681817000599e-05,
|
| 183 |
+
"loss": 0.3399,
|
| 184 |
+
"step": 14000
|
| 185 |
+
},
|
| 186 |
+
{
|
| 187 |
+
"epoch": 1.13,
|
| 188 |
+
"learning_rate": 1.2463813310464905e-05,
|
| 189 |
+
"loss": 0.3268,
|
| 190 |
+
"step": 14500
|
| 191 |
+
},
|
| 192 |
+
{
|
| 193 |
+
"epoch": 1.17,
|
| 194 |
+
"learning_rate": 1.2203944803929213e-05,
|
| 195 |
+
"loss": 0.3358,
|
| 196 |
+
"step": 15000
|
| 197 |
+
},
|
| 198 |
+
{
|
| 199 |
+
"epoch": 1.21,
|
| 200 |
+
"learning_rate": 1.1944076297393521e-05,
|
| 201 |
+
"loss": 0.3554,
|
| 202 |
+
"step": 15500
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 1.25,
|
| 206 |
+
"learning_rate": 1.1684207790857827e-05,
|
| 207 |
+
"loss": 0.337,
|
| 208 |
+
"step": 16000
|
| 209 |
+
},
|
| 210 |
+
{
|
| 211 |
+
"epoch": 1.29,
|
| 212 |
+
"learning_rate": 1.1424339284322135e-05,
|
| 213 |
+
"loss": 0.3321,
|
| 214 |
+
"step": 16500
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 1.33,
|
| 218 |
+
"learning_rate": 1.1164470777786441e-05,
|
| 219 |
+
"loss": 0.3314,
|
| 220 |
+
"step": 17000
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 1.36,
|
| 224 |
+
"learning_rate": 1.0904602271250749e-05,
|
| 225 |
+
"loss": 0.3243,
|
| 226 |
+
"step": 17500
|
| 227 |
+
},
|
| 228 |
+
{
|
| 229 |
+
"epoch": 1.4,
|
| 230 |
+
"learning_rate": 1.0644733764715057e-05,
|
| 231 |
+
"loss": 0.3147,
|
| 232 |
+
"step": 18000
|
| 233 |
+
},
|
| 234 |
+
{
|
| 235 |
+
"epoch": 1.44,
|
| 236 |
+
"learning_rate": 1.0384865258179361e-05,
|
| 237 |
+
"loss": 0.3184,
|
| 238 |
+
"step": 18500
|
| 239 |
+
},
|
| 240 |
+
{
|
| 241 |
+
"epoch": 1.48,
|
| 242 |
+
"learning_rate": 1.0124996751643669e-05,
|
| 243 |
+
"loss": 0.3638,
|
| 244 |
+
"step": 19000
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 1.52,
|
| 248 |
+
"learning_rate": 9.865128245107977e-06,
|
| 249 |
+
"loss": 0.3502,
|
| 250 |
+
"step": 19500
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 1.56,
|
| 254 |
+
"learning_rate": 9.605259738572283e-06,
|
| 255 |
+
"loss": 0.3486,
|
| 256 |
+
"step": 20000
|
| 257 |
+
},
|
| 258 |
+
{
|
| 259 |
+
"epoch": 1.6,
|
| 260 |
+
"learning_rate": 9.345391232036591e-06,
|
| 261 |
+
"loss": 0.388,
|
| 262 |
+
"step": 20500
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 1.64,
|
| 266 |
+
"learning_rate": 9.085522725500897e-06,
|
| 267 |
+
"loss": 0.4242,
|
| 268 |
+
"step": 21000
|
| 269 |
+
},
|
| 270 |
+
{
|
| 271 |
+
"epoch": 1.68,
|
| 272 |
+
"learning_rate": 8.825654218965203e-06,
|
| 273 |
+
"loss": 0.4053,
|
| 274 |
+
"step": 21500
|
| 275 |
+
},
|
| 276 |
+
{
|
| 277 |
+
"epoch": 1.72,
|
| 278 |
+
"learning_rate": 8.565785712429513e-06,
|
| 279 |
+
"loss": 0.3648,
|
| 280 |
+
"step": 22000
|
| 281 |
+
},
|
| 282 |
+
{
|
| 283 |
+
"epoch": 1.75,
|
| 284 |
+
"learning_rate": 8.305917205893819e-06,
|
| 285 |
+
"loss": 0.3675,
|
| 286 |
+
"step": 22500
|
| 287 |
+
},
|
| 288 |
+
{
|
| 289 |
+
"epoch": 1.79,
|
| 290 |
+
"learning_rate": 8.046048699358125e-06,
|
| 291 |
+
"loss": 0.3882,
|
| 292 |
+
"step": 23000
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 1.83,
|
| 296 |
+
"learning_rate": 7.786180192822433e-06,
|
| 297 |
+
"loss": 0.4318,
|
| 298 |
+
"step": 23500
|
| 299 |
+
},
|
| 300 |
+
{
|
| 301 |
+
"epoch": 1.87,
|
| 302 |
+
"learning_rate": 7.526311686286739e-06,
|
| 303 |
+
"loss": 0.402,
|
| 304 |
+
"step": 24000
|
| 305 |
+
},
|
| 306 |
+
{
|
| 307 |
+
"epoch": 1.91,
|
| 308 |
+
"learning_rate": 7.266443179751047e-06,
|
| 309 |
+
"loss": 0.3708,
|
| 310 |
+
"step": 24500
|
| 311 |
+
},
|
| 312 |
+
{
|
| 313 |
+
"epoch": 1.95,
|
| 314 |
+
"learning_rate": 7.006574673215354e-06,
|
| 315 |
+
"loss": 0.3576,
|
| 316 |
+
"step": 25000
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 1.99,
|
| 320 |
+
"learning_rate": 6.746706166679661e-06,
|
| 321 |
+
"loss": 0.3417,
|
| 322 |
+
"step": 25500
|
| 323 |
+
},
|
| 324 |
+
{
|
| 325 |
+
"epoch": 2.0,
|
| 326 |
+
"eval_loss": 0.3679686188697815,
|
| 327 |
+
"eval_runtime": 658.8287,
|
| 328 |
+
"eval_samples_per_second": 66.752,
|
| 329 |
+
"eval_steps_per_second": 8.345,
|
| 330 |
+
"step": 25654
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 2.03,
|
| 334 |
+
"learning_rate": 6.486837660143968e-06,
|
| 335 |
+
"loss": 0.355,
|
| 336 |
+
"step": 26000
|
| 337 |
+
},
|
| 338 |
+
{
|
| 339 |
+
"epoch": 2.07,
|
| 340 |
+
"learning_rate": 6.226969153608275e-06,
|
| 341 |
+
"loss": 0.3088,
|
| 342 |
+
"step": 26500
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 2.1,
|
| 346 |
+
"learning_rate": 5.967100647072583e-06,
|
| 347 |
+
"loss": 0.296,
|
| 348 |
+
"step": 27000
|
| 349 |
+
},
|
| 350 |
+
{
|
| 351 |
+
"epoch": 2.14,
|
| 352 |
+
"learning_rate": 5.707232140536889e-06,
|
| 353 |
+
"loss": 0.2952,
|
| 354 |
+
"step": 27500
|
| 355 |
+
},
|
| 356 |
+
{
|
| 357 |
+
"epoch": 2.18,
|
| 358 |
+
"learning_rate": 5.447363634001196e-06,
|
| 359 |
+
"loss": 0.2896,
|
| 360 |
+
"step": 28000
|
| 361 |
+
},
|
| 362 |
+
{
|
| 363 |
+
"epoch": 2.22,
|
| 364 |
+
"learning_rate": 5.187495127465503e-06,
|
| 365 |
+
"loss": 0.2768,
|
| 366 |
+
"step": 28500
|
| 367 |
+
},
|
| 368 |
+
{
|
| 369 |
+
"epoch": 2.26,
|
| 370 |
+
"learning_rate": 4.9276266209298106e-06,
|
| 371 |
+
"loss": 0.266,
|
| 372 |
+
"step": 29000
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 2.3,
|
| 376 |
+
"learning_rate": 4.667758114394117e-06,
|
| 377 |
+
"loss": 0.2753,
|
| 378 |
+
"step": 29500
|
| 379 |
+
},
|
| 380 |
+
{
|
| 381 |
+
"epoch": 2.34,
|
| 382 |
+
"learning_rate": 4.407889607858424e-06,
|
| 383 |
+
"loss": 0.2773,
|
| 384 |
+
"step": 30000
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 2.38,
|
| 388 |
+
"learning_rate": 4.1480211013227315e-06,
|
| 389 |
+
"loss": 0.2755,
|
| 390 |
+
"step": 30500
|
| 391 |
+
},
|
| 392 |
+
{
|
| 393 |
+
"epoch": 2.42,
|
| 394 |
+
"learning_rate": 3.888152594787038e-06,
|
| 395 |
+
"loss": 0.27,
|
| 396 |
+
"step": 31000
|
| 397 |
+
},
|
| 398 |
+
{
|
| 399 |
+
"epoch": 2.46,
|
| 400 |
+
"learning_rate": 3.6282840882513455e-06,
|
| 401 |
+
"loss": 0.2683,
|
| 402 |
+
"step": 31500
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 2.49,
|
| 406 |
+
"learning_rate": 3.368415581715652e-06,
|
| 407 |
+
"loss": 0.2718,
|
| 408 |
+
"step": 32000
|
| 409 |
+
},
|
| 410 |
+
{
|
| 411 |
+
"epoch": 2.53,
|
| 412 |
+
"learning_rate": 3.108547075179959e-06,
|
| 413 |
+
"loss": 0.2824,
|
| 414 |
+
"step": 32500
|
| 415 |
+
},
|
| 416 |
+
{
|
| 417 |
+
"epoch": 2.57,
|
| 418 |
+
"learning_rate": 2.8486785686442665e-06,
|
| 419 |
+
"loss": 0.2691,
|
| 420 |
+
"step": 33000
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 2.61,
|
| 424 |
+
"learning_rate": 2.588810062108573e-06,
|
| 425 |
+
"loss": 0.2664,
|
| 426 |
+
"step": 33500
|
| 427 |
+
},
|
| 428 |
+
{
|
| 429 |
+
"epoch": 2.65,
|
| 430 |
+
"learning_rate": 2.3289415555728805e-06,
|
| 431 |
+
"loss": 0.2655,
|
| 432 |
+
"step": 34000
|
| 433 |
+
},
|
| 434 |
+
{
|
| 435 |
+
"epoch": 2.69,
|
| 436 |
+
"learning_rate": 2.0690730490371875e-06,
|
| 437 |
+
"loss": 0.2681,
|
| 438 |
+
"step": 34500
|
| 439 |
+
},
|
| 440 |
+
{
|
| 441 |
+
"epoch": 2.73,
|
| 442 |
+
"learning_rate": 1.8092045425014945e-06,
|
| 443 |
+
"loss": 0.2751,
|
| 444 |
+
"step": 35000
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 2.77,
|
| 448 |
+
"learning_rate": 1.5493360359658013e-06,
|
| 449 |
+
"loss": 0.2731,
|
| 450 |
+
"step": 35500
|
| 451 |
+
},
|
| 452 |
+
{
|
| 453 |
+
"epoch": 2.81,
|
| 454 |
+
"learning_rate": 1.2894675294301085e-06,
|
| 455 |
+
"loss": 0.2642,
|
| 456 |
+
"step": 36000
|
| 457 |
+
},
|
| 458 |
+
{
|
| 459 |
+
"epoch": 2.85,
|
| 460 |
+
"learning_rate": 1.0295990228944154e-06,
|
| 461 |
+
"loss": 0.2622,
|
| 462 |
+
"step": 36500
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 2.88,
|
| 466 |
+
"learning_rate": 7.697305163587225e-07,
|
| 467 |
+
"loss": 0.2716,
|
| 468 |
+
"step": 37000
|
| 469 |
+
},
|
| 470 |
+
{
|
| 471 |
+
"epoch": 2.92,
|
| 472 |
+
"learning_rate": 5.098620098230296e-07,
|
| 473 |
+
"loss": 0.2717,
|
| 474 |
+
"step": 37500
|
| 475 |
+
},
|
| 476 |
+
{
|
| 477 |
+
"epoch": 2.96,
|
| 478 |
+
"learning_rate": 2.4999350328733664e-07,
|
| 479 |
+
"loss": 0.2672,
|
| 480 |
+
"step": 38000
|
| 481 |
+
}
|
| 482 |
+
],
|
| 483 |
+
"max_steps": 38481,
|
| 484 |
+
"num_train_epochs": 3,
|
| 485 |
+
"total_flos": 3.5290626392064e+16,
|
| 486 |
+
"trial_name": null,
|
| 487 |
+
"trial_params": null
|
| 488 |
+
}
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:145c38e2eb782ce5b677946aa11453aa06daf69ccbc6394a571eca5e8e065f74
|
| 3 |
+
size 3451
|