Training in progress, step 3000, checkpoint
Browse files
checkpoint-3000/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"annotation_transformer": false,
|
| 3 |
+
"architectures": [
|
| 4 |
+
"CAMP"
|
| 5 |
+
],
|
| 6 |
+
"diff": false,
|
| 7 |
+
"dropout": 0.05,
|
| 8 |
+
"hidden_dim": 640,
|
| 9 |
+
"input_dim": 384,
|
| 10 |
+
"intermediate_dim": 2560,
|
| 11 |
+
"kernel_size": 11,
|
| 12 |
+
"latent": false,
|
| 13 |
+
"mlm": false,
|
| 14 |
+
"mnr": false,
|
| 15 |
+
"model_type": "CAMP",
|
| 16 |
+
"nhead": 8,
|
| 17 |
+
"nlp_path": "allenai/scibert_scivocab_uncased",
|
| 18 |
+
"num_hidden_layers": 1,
|
| 19 |
+
"out_dim": 512,
|
| 20 |
+
"plm_path": "facebook/esm2_t33_650m_UR50D",
|
| 21 |
+
"pooling": "avg",
|
| 22 |
+
"token": null,
|
| 23 |
+
"torch_dtype": "bfloat16",
|
| 24 |
+
"transformers_version": "4.41.2"
|
| 25 |
+
}
|
checkpoint-3000/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4f500b359798b004b50fa98053fb5e828466ddff777e249ba12bfce12c996bf6
|
| 3 |
+
size 1570427506
|
checkpoint-3000/trainer_state.json
ADDED
|
@@ -0,0 +1,276 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.0671952694530305,
|
| 5 |
+
"eval_steps": 1000,
|
| 6 |
+
"global_step": 3000,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.002239842315101017,
|
| 13 |
+
"grad_norm": 71.10852813720703,
|
| 14 |
+
"learning_rate": 0.0001,
|
| 15 |
+
"loss": 3.5057,
|
| 16 |
+
"step": 100
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.004479684630202034,
|
| 20 |
+
"grad_norm": 57.756534576416016,
|
| 21 |
+
"learning_rate": 9.9998756572327e-05,
|
| 22 |
+
"loss": 0.8254,
|
| 23 |
+
"step": 200
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.006719526945303051,
|
| 27 |
+
"grad_norm": 44.61080551147461,
|
| 28 |
+
"learning_rate": 9.999502635115246e-05,
|
| 29 |
+
"loss": 0.6935,
|
| 30 |
+
"step": 300
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.008959369260404068,
|
| 34 |
+
"grad_norm": 50.82619094848633,
|
| 35 |
+
"learning_rate": 9.998880952200681e-05,
|
| 36 |
+
"loss": 0.5972,
|
| 37 |
+
"step": 400
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.011199211575505085,
|
| 41 |
+
"grad_norm": 44.26677703857422,
|
| 42 |
+
"learning_rate": 9.998010639409713e-05,
|
| 43 |
+
"loss": 0.5408,
|
| 44 |
+
"step": 500
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.013439053890606102,
|
| 48 |
+
"grad_norm": 42.8045654296875,
|
| 49 |
+
"learning_rate": 9.996891740029186e-05,
|
| 50 |
+
"loss": 0.5509,
|
| 51 |
+
"step": 600
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.01567889620570712,
|
| 55 |
+
"grad_norm": 30.989139556884766,
|
| 56 |
+
"learning_rate": 9.995524309709913e-05,
|
| 57 |
+
"loss": 0.4823,
|
| 58 |
+
"step": 700
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.017918738520808136,
|
| 62 |
+
"grad_norm": 34.01952362060547,
|
| 63 |
+
"learning_rate": 9.993908416463927e-05,
|
| 64 |
+
"loss": 0.5111,
|
| 65 |
+
"step": 800
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.020158580835909153,
|
| 69 |
+
"grad_norm": 34.079307556152344,
|
| 70 |
+
"learning_rate": 9.992044140661079e-05,
|
| 71 |
+
"loss": 0.4635,
|
| 72 |
+
"step": 900
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.02239842315101017,
|
| 76 |
+
"grad_norm": 26.16071128845215,
|
| 77 |
+
"learning_rate": 9.989931575025056e-05,
|
| 78 |
+
"loss": 0.4883,
|
| 79 |
+
"step": 1000
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.02239842315101017,
|
| 83 |
+
"eval_avg_non_pair_similarity": 0.0020343252948339737,
|
| 84 |
+
"eval_avg_pair_similarity": 0.008852629057131708,
|
| 85 |
+
"eval_loss": 0.5431402921676636,
|
| 86 |
+
"eval_runtime": 19.6178,
|
| 87 |
+
"eval_samples_per_second": 25.487,
|
| 88 |
+
"eval_similarity_ratio": 4.35162905343228,
|
| 89 |
+
"eval_steps_per_second": 0.816,
|
| 90 |
+
"step": 1000
|
| 91 |
+
},
|
| 92 |
+
{
|
| 93 |
+
"epoch": 0.024638265466111187,
|
| 94 |
+
"grad_norm": 35.38695526123047,
|
| 95 |
+
"learning_rate": 9.987570824628759e-05,
|
| 96 |
+
"loss": 0.4655,
|
| 97 |
+
"step": 1100
|
| 98 |
+
},
|
| 99 |
+
{
|
| 100 |
+
"epoch": 0.026878107781212204,
|
| 101 |
+
"grad_norm": 32.071346282958984,
|
| 102 |
+
"learning_rate": 9.984962006889084e-05,
|
| 103 |
+
"loss": 0.4342,
|
| 104 |
+
"step": 1200
|
| 105 |
+
},
|
| 106 |
+
{
|
| 107 |
+
"epoch": 0.029117950096313218,
|
| 108 |
+
"grad_norm": 39.6610221862793,
|
| 109 |
+
"learning_rate": 9.982105251561082e-05,
|
| 110 |
+
"loss": 0.458,
|
| 111 |
+
"step": 1300
|
| 112 |
+
},
|
| 113 |
+
{
|
| 114 |
+
"epoch": 0.03135779241141424,
|
| 115 |
+
"grad_norm": 31.493322372436523,
|
| 116 |
+
"learning_rate": 9.979000700731491e-05,
|
| 117 |
+
"loss": 0.4525,
|
| 118 |
+
"step": 1400
|
| 119 |
+
},
|
| 120 |
+
{
|
| 121 |
+
"epoch": 0.03359763472651525,
|
| 122 |
+
"grad_norm": 34.453399658203125,
|
| 123 |
+
"learning_rate": 9.975648508811693e-05,
|
| 124 |
+
"loss": 0.41,
|
| 125 |
+
"step": 1500
|
| 126 |
+
},
|
| 127 |
+
{
|
| 128 |
+
"epoch": 0.03583747704161627,
|
| 129 |
+
"grad_norm": 33.990074157714844,
|
| 130 |
+
"learning_rate": 9.972048842530012e-05,
|
| 131 |
+
"loss": 0.4097,
|
| 132 |
+
"step": 1600
|
| 133 |
+
},
|
| 134 |
+
{
|
| 135 |
+
"epoch": 0.038077319356717286,
|
| 136 |
+
"grad_norm": 30.44228172302246,
|
| 137 |
+
"learning_rate": 9.968201880923439e-05,
|
| 138 |
+
"loss": 0.4257,
|
| 139 |
+
"step": 1700
|
| 140 |
+
},
|
| 141 |
+
{
|
| 142 |
+
"epoch": 0.040317161671818307,
|
| 143 |
+
"grad_norm": 31.427162170410156,
|
| 144 |
+
"learning_rate": 9.964107815328711e-05,
|
| 145 |
+
"loss": 0.3821,
|
| 146 |
+
"step": 1800
|
| 147 |
+
},
|
| 148 |
+
{
|
| 149 |
+
"epoch": 0.04255700398691932,
|
| 150 |
+
"grad_norm": 25.877887725830078,
|
| 151 |
+
"learning_rate": 9.959766849372808e-05,
|
| 152 |
+
"loss": 0.3788,
|
| 153 |
+
"step": 1900
|
| 154 |
+
},
|
| 155 |
+
{
|
| 156 |
+
"epoch": 0.04479684630202034,
|
| 157 |
+
"grad_norm": 25.36798095703125,
|
| 158 |
+
"learning_rate": 9.955179198962817e-05,
|
| 159 |
+
"loss": 0.3854,
|
| 160 |
+
"step": 2000
|
| 161 |
+
},
|
| 162 |
+
{
|
| 163 |
+
"epoch": 0.04479684630202034,
|
| 164 |
+
"eval_avg_non_pair_similarity": 0.0016289287904792565,
|
| 165 |
+
"eval_avg_pair_similarity": 0.0032495629731565715,
|
| 166 |
+
"eval_loss": 0.4394480586051941,
|
| 167 |
+
"eval_runtime": 19.5064,
|
| 168 |
+
"eval_samples_per_second": 25.633,
|
| 169 |
+
"eval_similarity_ratio": 1.9949079371360972,
|
| 170 |
+
"eval_steps_per_second": 0.82,
|
| 171 |
+
"step": 2000
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.047036688617121354,
|
| 175 |
+
"grad_norm": 24.537384033203125,
|
| 176 |
+
"learning_rate": 9.950345092275198e-05,
|
| 177 |
+
"loss": 0.3886,
|
| 178 |
+
"step": 2100
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.049276530932222375,
|
| 182 |
+
"grad_norm": 26.4981632232666,
|
| 183 |
+
"learning_rate": 9.945264769744431e-05,
|
| 184 |
+
"loss": 0.3926,
|
| 185 |
+
"step": 2200
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.05151637324732339,
|
| 189 |
+
"grad_norm": 40.585941314697266,
|
| 190 |
+
"learning_rate": 9.939938484051063e-05,
|
| 191 |
+
"loss": 0.3796,
|
| 192 |
+
"step": 2300
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.05375621556242441,
|
| 196 |
+
"grad_norm": 26.374149322509766,
|
| 197 |
+
"learning_rate": 9.934366500109132e-05,
|
| 198 |
+
"loss": 0.3723,
|
| 199 |
+
"step": 2400
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.05599605787752542,
|
| 203 |
+
"grad_norm": 31.54728126525879,
|
| 204 |
+
"learning_rate": 9.928549095053001e-05,
|
| 205 |
+
"loss": 0.3639,
|
| 206 |
+
"step": 2500
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.058235900192626436,
|
| 210 |
+
"grad_norm": 26.00455093383789,
|
| 211 |
+
"learning_rate": 9.922486558223567e-05,
|
| 212 |
+
"loss": 0.3387,
|
| 213 |
+
"step": 2600
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.060475742507727456,
|
| 217 |
+
"grad_norm": 23.684057235717773,
|
| 218 |
+
"learning_rate": 9.916179191153873e-05,
|
| 219 |
+
"loss": 0.3494,
|
| 220 |
+
"step": 2700
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.06271558482282848,
|
| 224 |
+
"grad_norm": 24.034656524658203,
|
| 225 |
+
"learning_rate": 9.909627307554108e-05,
|
| 226 |
+
"loss": 0.362,
|
| 227 |
+
"step": 2800
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.06495542713792948,
|
| 231 |
+
"grad_norm": 31.07891082763672,
|
| 232 |
+
"learning_rate": 9.902831233296009e-05,
|
| 233 |
+
"loss": 0.3481,
|
| 234 |
+
"step": 2900
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.0671952694530305,
|
| 238 |
+
"grad_norm": 24.183626174926758,
|
| 239 |
+
"learning_rate": 9.895791306396644e-05,
|
| 240 |
+
"loss": 0.3535,
|
| 241 |
+
"step": 3000
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.0671952694530305,
|
| 245 |
+
"eval_avg_non_pair_similarity": -0.0007878901720614375,
|
| 246 |
+
"eval_avg_pair_similarity": -0.008789425778668375,
|
| 247 |
+
"eval_loss": 0.4187028110027313,
|
| 248 |
+
"eval_runtime": 19.4839,
|
| 249 |
+
"eval_samples_per_second": 25.662,
|
| 250 |
+
"eval_similarity_ratio": 11.155648452463497,
|
| 251 |
+
"eval_steps_per_second": 0.821,
|
| 252 |
+
"step": 3000
|
| 253 |
+
}
|
| 254 |
+
],
|
| 255 |
+
"logging_steps": 100,
|
| 256 |
+
"max_steps": 44646,
|
| 257 |
+
"num_input_tokens_seen": 0,
|
| 258 |
+
"num_train_epochs": 1,
|
| 259 |
+
"save_steps": 1000,
|
| 260 |
+
"stateful_callbacks": {
|
| 261 |
+
"TrainerControl": {
|
| 262 |
+
"args": {
|
| 263 |
+
"should_epoch_stop": false,
|
| 264 |
+
"should_evaluate": false,
|
| 265 |
+
"should_log": false,
|
| 266 |
+
"should_save": true,
|
| 267 |
+
"should_training_stop": false
|
| 268 |
+
},
|
| 269 |
+
"attributes": {}
|
| 270 |
+
}
|
| 271 |
+
},
|
| 272 |
+
"total_flos": 0.0,
|
| 273 |
+
"train_batch_size": 32,
|
| 274 |
+
"trial_name": null,
|
| 275 |
+
"trial_params": null
|
| 276 |
+
}
|
checkpoint-3000/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7c6bbe0930b5977d5ef83c9c6a2872aea787e9f64c3d133a00319e43c833ad34
|
| 3 |
+
size 5112
|