Upload folder using huggingface_hub
Browse files
shared/checkpoints/latest/config.json
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"AxaiForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_dropout": 0.0,
|
| 6 |
+
"dtype": "float32",
|
| 7 |
+
"hidden_dropout": 0.0,
|
| 8 |
+
"hidden_size": 768,
|
| 9 |
+
"initializer_range": 0.02,
|
| 10 |
+
"intermediate_size": 3072,
|
| 11 |
+
"max_position_embeddings": 128,
|
| 12 |
+
"model_type": "axai",
|
| 13 |
+
"neftune_alpha": 0.0,
|
| 14 |
+
"num_attention_heads": 12,
|
| 15 |
+
"num_hidden_layers": 24,
|
| 16 |
+
"num_key_value_heads": 6,
|
| 17 |
+
"qk_norm": true,
|
| 18 |
+
"rezero_init": 1.0,
|
| 19 |
+
"rms_norm_eps": 1e-06,
|
| 20 |
+
"rope_theta": 10000.0,
|
| 21 |
+
"transformers_version": "5.0.0",
|
| 22 |
+
"use_cache": false,
|
| 23 |
+
"vocab_size": 32000
|
| 24 |
+
}
|
shared/checkpoints/latest/trainer_state.json
ADDED
|
@@ -0,0 +1,321 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_global_step": null,
|
| 3 |
+
"best_metric": null,
|
| 4 |
+
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.9432293815952367,
|
| 6 |
+
"eval_steps": 500,
|
| 7 |
+
"global_step": 800,
|
| 8 |
+
"is_hyper_param_search": false,
|
| 9 |
+
"is_local_process_zero": true,
|
| 10 |
+
"is_world_process_zero": true,
|
| 11 |
+
"log_history": [
|
| 12 |
+
{
|
| 13 |
+
"epoch": 0.001179036726994046,
|
| 14 |
+
"grad_norm": 18.818986892700195,
|
| 15 |
+
"learning_rate": 5e-05,
|
| 16 |
+
"loss": 214.36204528808594,
|
| 17 |
+
"step": 1
|
| 18 |
+
},
|
| 19 |
+
{
|
| 20 |
+
"epoch": 0.023580734539880917,
|
| 21 |
+
"grad_norm": 17.1884765625,
|
| 22 |
+
"learning_rate": 4.993823781115963e-05,
|
| 23 |
+
"loss": 199.185546875,
|
| 24 |
+
"step": 20
|
| 25 |
+
},
|
| 26 |
+
{
|
| 27 |
+
"epoch": 0.047161469079761834,
|
| 28 |
+
"grad_norm": 18.005523681640625,
|
| 29 |
+
"learning_rate": 4.97401218720448e-05,
|
| 30 |
+
"loss": 203.69400634765626,
|
| 31 |
+
"step": 40
|
| 32 |
+
},
|
| 33 |
+
{
|
| 34 |
+
"epoch": 0.07074220361964276,
|
| 35 |
+
"grad_norm": 17.965116500854492,
|
| 36 |
+
"learning_rate": 4.940656561583624e-05,
|
| 37 |
+
"loss": 199.9952880859375,
|
| 38 |
+
"step": 60
|
| 39 |
+
},
|
| 40 |
+
{
|
| 41 |
+
"epoch": 0.09432293815952367,
|
| 42 |
+
"grad_norm": 17.385486602783203,
|
| 43 |
+
"learning_rate": 4.893939510326814e-05,
|
| 44 |
+
"loss": 200.97593994140624,
|
| 45 |
+
"step": 80
|
| 46 |
+
},
|
| 47 |
+
{
|
| 48 |
+
"epoch": 0.11790367269940459,
|
| 49 |
+
"grad_norm": 18.995485305786133,
|
| 50 |
+
"learning_rate": 4.834116786912897e-05,
|
| 51 |
+
"loss": 197.9327880859375,
|
| 52 |
+
"step": 100
|
| 53 |
+
},
|
| 54 |
+
{
|
| 55 |
+
"epoch": 0.14148440723928551,
|
| 56 |
+
"grad_norm": 17.99912452697754,
|
| 57 |
+
"learning_rate": 4.761515892098357e-05,
|
| 58 |
+
"loss": 200.61953125,
|
| 59 |
+
"step": 120
|
| 60 |
+
},
|
| 61 |
+
{
|
| 62 |
+
"epoch": 0.16506514177916642,
|
| 63 |
+
"grad_norm": 18.637039184570312,
|
| 64 |
+
"learning_rate": 4.67653428100754e-05,
|
| 65 |
+
"loss": 195.87960205078124,
|
| 66 |
+
"step": 140
|
| 67 |
+
},
|
| 68 |
+
{
|
| 69 |
+
"epoch": 0.18864587631904733,
|
| 70 |
+
"grad_norm": 19.77713966369629,
|
| 71 |
+
"learning_rate": 4.579637187256222e-05,
|
| 72 |
+
"loss": 203.71815185546876,
|
| 73 |
+
"step": 160
|
| 74 |
+
},
|
| 75 |
+
{
|
| 76 |
+
"epoch": 0.21222661085892824,
|
| 77 |
+
"grad_norm": 20.302921295166016,
|
| 78 |
+
"learning_rate": 4.4713550760204035e-05,
|
| 79 |
+
"loss": 201.46798095703124,
|
| 80 |
+
"step": 180
|
| 81 |
+
},
|
| 82 |
+
{
|
| 83 |
+
"epoch": 0.23580734539880918,
|
| 84 |
+
"grad_norm": 20.111055374145508,
|
| 85 |
+
"learning_rate": 4.352280739993557e-05,
|
| 86 |
+
"loss": 191.88206787109374,
|
| 87 |
+
"step": 200
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.25938807993869006,
|
| 91 |
+
"grad_norm": 20.618942260742188,
|
| 92 |
+
"learning_rate": 4.223066054130568e-05,
|
| 93 |
+
"loss": 189.92415771484374,
|
| 94 |
+
"step": 220
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.28296881447857103,
|
| 98 |
+
"grad_norm": 20.38855743408203,
|
| 99 |
+
"learning_rate": 4.0844184069445955e-05,
|
| 100 |
+
"loss": 197.627880859375,
|
| 101 |
+
"step": 240
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.30654954901845194,
|
| 105 |
+
"grad_norm": 22.40736198425293,
|
| 106 |
+
"learning_rate": 3.937096827893793e-05,
|
| 107 |
+
"loss": 192.96744384765626,
|
| 108 |
+
"step": 260
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.33013028355833285,
|
| 112 |
+
"grad_norm": 26.542959213256836,
|
| 113 |
+
"learning_rate": 3.781907832058587e-05,
|
| 114 |
+
"loss": 192.7069580078125,
|
| 115 |
+
"step": 280
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.35371101809821376,
|
| 119 |
+
"grad_norm": 24.460668563842773,
|
| 120 |
+
"learning_rate": 3.619701004857919e-05,
|
| 121 |
+
"loss": 193.03927001953124,
|
| 122 |
+
"step": 300
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.37729175263809467,
|
| 126 |
+
"grad_norm": 22.90284538269043,
|
| 127 |
+
"learning_rate": 3.451364350976022e-05,
|
| 128 |
+
"loss": 185.04234619140624,
|
| 129 |
+
"step": 320
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.4008724871779756,
|
| 133 |
+
"grad_norm": 23.723445892333984,
|
| 134 |
+
"learning_rate": 3.2778194329621104e-05,
|
| 135 |
+
"loss": 190.406298828125,
|
| 136 |
+
"step": 340
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.4244532217178565,
|
| 140 |
+
"grad_norm": 25.815519332885742,
|
| 141 |
+
"learning_rate": 3.1000163261168366e-05,
|
| 142 |
+
"loss": 188.7949951171875,
|
| 143 |
+
"step": 360
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.44803395625773745,
|
| 147 |
+
"grad_norm": 26.482738494873047,
|
| 148 |
+
"learning_rate": 2.9189284172850983e-05,
|
| 149 |
+
"loss": 187.61990966796876,
|
| 150 |
+
"step": 380
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.47161469079761836,
|
| 154 |
+
"grad_norm": 25.560789108276367,
|
| 155 |
+
"learning_rate": 2.7355470760292956e-05,
|
| 156 |
+
"loss": 188.86490478515626,
|
| 157 |
+
"step": 400
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.4951954253374993,
|
| 161 |
+
"grad_norm": 26.315908432006836,
|
| 162 |
+
"learning_rate": 2.5508762273558284e-05,
|
| 163 |
+
"loss": 190.40867919921874,
|
| 164 |
+
"step": 420
|
| 165 |
+
},
|
| 166 |
+
{
|
| 167 |
+
"epoch": 0.5187761598773801,
|
| 168 |
+
"grad_norm": 23.34573745727539,
|
| 169 |
+
"learning_rate": 2.3659268557065544e-05,
|
| 170 |
+
"loss": 191.79708251953124,
|
| 171 |
+
"step": 440
|
| 172 |
+
},
|
| 173 |
+
{
|
| 174 |
+
"epoch": 0.5423568944172611,
|
| 175 |
+
"grad_norm": 26.028112411499023,
|
| 176 |
+
"learning_rate": 2.1817114703032176e-05,
|
| 177 |
+
"loss": 188.81153564453126,
|
| 178 |
+
"step": 460
|
| 179 |
+
},
|
| 180 |
+
{
|
| 181 |
+
"epoch": 0.5659376289571421,
|
| 182 |
+
"grad_norm": 26.664793014526367,
|
| 183 |
+
"learning_rate": 1.99923856214443e-05,
|
| 184 |
+
"loss": 182.7518798828125,
|
| 185 |
+
"step": 480
|
| 186 |
+
},
|
| 187 |
+
{
|
| 188 |
+
"epoch": 0.5895183634970229,
|
| 189 |
+
"grad_norm": 25.86208152770996,
|
| 190 |
+
"learning_rate": 1.819507083000514e-05,
|
| 191 |
+
"loss": 189.74847412109375,
|
| 192 |
+
"step": 500
|
| 193 |
+
},
|
| 194 |
+
{
|
| 195 |
+
"epoch": 0.6130990980369039,
|
| 196 |
+
"grad_norm": 26.640764236450195,
|
| 197 |
+
"learning_rate": 1.643500976631037e-05,
|
| 198 |
+
"loss": 186.2022216796875,
|
| 199 |
+
"step": 520
|
| 200 |
+
},
|
| 201 |
+
{
|
| 202 |
+
"epoch": 0.6366798325767847,
|
| 203 |
+
"grad_norm": 29.260278701782227,
|
| 204 |
+
"learning_rate": 1.47218379216403e-05,
|
| 205 |
+
"loss": 186.85718994140626,
|
| 206 |
+
"step": 540
|
| 207 |
+
},
|
| 208 |
+
{
|
| 209 |
+
"epoch": 0.6602605671166657,
|
| 210 |
+
"grad_norm": 25.371997833251953,
|
| 211 |
+
"learning_rate": 1.3064934091260262e-05,
|
| 212 |
+
"loss": 190.2005126953125,
|
| 213 |
+
"step": 560
|
| 214 |
+
},
|
| 215 |
+
{
|
| 216 |
+
"epoch": 0.6838413016565466,
|
| 217 |
+
"grad_norm": 23.361217498779297,
|
| 218 |
+
"learning_rate": 1.1473369030008974e-05,
|
| 219 |
+
"loss": 192.95072021484376,
|
| 220 |
+
"step": 580
|
| 221 |
+
},
|
| 222 |
+
{
|
| 223 |
+
"epoch": 0.7074220361964275,
|
| 224 |
+
"grad_norm": 26.195369720458984,
|
| 225 |
+
"learning_rate": 9.955855794260422e-06,
|
| 226 |
+
"loss": 187.2462646484375,
|
| 227 |
+
"step": 600
|
| 228 |
+
},
|
| 229 |
+
{
|
| 230 |
+
"epoch": 0.7310027707363085,
|
| 231 |
+
"grad_norm": 27.924829483032227,
|
| 232 |
+
"learning_rate": 8.520702042113765e-06,
|
| 233 |
+
"loss": 187.06983642578126,
|
| 234 |
+
"step": 620
|
| 235 |
+
},
|
| 236 |
+
{
|
| 237 |
+
"epoch": 0.7545835052761893,
|
| 238 |
+
"grad_norm": 25.70247459411621,
|
| 239 |
+
"learning_rate": 7.1757645529443665e-06,
|
| 240 |
+
"loss": 178.98265380859374,
|
| 241 |
+
"step": 640
|
| 242 |
+
},
|
| 243 |
+
{
|
| 244 |
+
"epoch": 0.7781642398160703,
|
| 245 |
+
"grad_norm": 25.393587112426758,
|
| 246 |
+
"learning_rate": 5.928406215300114e-06,
|
| 247 |
+
"loss": 186.89532470703125,
|
| 248 |
+
"step": 660
|
| 249 |
+
},
|
| 250 |
+
{
|
| 251 |
+
"epoch": 0.8017449743559512,
|
| 252 |
+
"grad_norm": 26.78150177001953,
|
| 253 |
+
"learning_rate": 4.785455718613227e-06,
|
| 254 |
+
"loss": 181.5225341796875,
|
| 255 |
+
"step": 680
|
| 256 |
+
},
|
| 257 |
+
{
|
| 258 |
+
"epoch": 0.8253257088958321,
|
| 259 |
+
"grad_norm": 26.85440444946289,
|
| 260 |
+
"learning_rate": 3.7531701693965554e-06,
|
| 261 |
+
"loss": 180.88773193359376,
|
| 262 |
+
"step": 700
|
| 263 |
+
},
|
| 264 |
+
{
|
| 265 |
+
"epoch": 0.848906443435713,
|
| 266 |
+
"grad_norm": 28.118528366088867,
|
| 267 |
+
"learning_rate": 2.8372008365823904e-06,
|
| 268 |
+
"loss": 183.96854248046876,
|
| 269 |
+
"step": 720
|
| 270 |
+
},
|
| 271 |
+
{
|
| 272 |
+
"epoch": 0.8724871779755939,
|
| 273 |
+
"grad_norm": 26.2855167388916,
|
| 274 |
+
"learning_rate": 2.0425622135320262e-06,
|
| 275 |
+
"loss": 183.86905517578126,
|
| 276 |
+
"step": 740
|
| 277 |
+
},
|
| 278 |
+
{
|
| 279 |
+
"epoch": 0.8960679125154749,
|
| 280 |
+
"grad_norm": 25.188119888305664,
|
| 281 |
+
"learning_rate": 1.3736045660864034e-06,
|
| 282 |
+
"loss": 186.562548828125,
|
| 283 |
+
"step": 760
|
| 284 |
+
},
|
| 285 |
+
{
|
| 286 |
+
"epoch": 0.9196486470553558,
|
| 287 |
+
"grad_norm": 26.751577377319336,
|
| 288 |
+
"learning_rate": 8.339901169443831e-07,
|
| 289 |
+
"loss": 186.4565673828125,
|
| 290 |
+
"step": 780
|
| 291 |
+
},
|
| 292 |
+
{
|
| 293 |
+
"epoch": 0.9432293815952367,
|
| 294 |
+
"grad_norm": 25.48533821105957,
|
| 295 |
+
"learning_rate": 4.266729967476013e-07,
|
| 296 |
+
"loss": 185.49075927734376,
|
| 297 |
+
"step": 800
|
| 298 |
+
}
|
| 299 |
+
],
|
| 300 |
+
"logging_steps": 20,
|
| 301 |
+
"max_steps": 849,
|
| 302 |
+
"num_input_tokens_seen": 0,
|
| 303 |
+
"num_train_epochs": 1,
|
| 304 |
+
"save_steps": 400,
|
| 305 |
+
"stateful_callbacks": {
|
| 306 |
+
"TrainerControl": {
|
| 307 |
+
"args": {
|
| 308 |
+
"should_epoch_stop": false,
|
| 309 |
+
"should_evaluate": false,
|
| 310 |
+
"should_log": false,
|
| 311 |
+
"should_save": true,
|
| 312 |
+
"should_training_stop": false
|
| 313 |
+
},
|
| 314 |
+
"attributes": {}
|
| 315 |
+
}
|
| 316 |
+
},
|
| 317 |
+
"total_flos": 1.234689785856e+16,
|
| 318 |
+
"train_batch_size": 1,
|
| 319 |
+
"trial_name": null,
|
| 320 |
+
"trial_params": null
|
| 321 |
+
}
|