Training in progress, step 500, checkpoint
Browse files- last-checkpoint/config.json +4 -4
- last-checkpoint/model.safetensors +2 -2
- last-checkpoint/optimizer.pt +2 -2
- last-checkpoint/rng_state_0.pth +1 -1
- last-checkpoint/rng_state_1.pth +1 -1
- last-checkpoint/rng_state_2.pth +1 -1
- last-checkpoint/rng_state_3.pth +1 -1
- last-checkpoint/scheduler.pt +1 -1
- last-checkpoint/trainer_state.json +23 -513
- last-checkpoint/training_args.bin +1 -1
last-checkpoint/config.json
CHANGED
|
@@ -1,5 +1,5 @@
|
|
| 1 |
{
|
| 2 |
-
"_name_or_path": "tattabio/
|
| 3 |
"architectures": [
|
| 4 |
"gLM2ForMaskedLM"
|
| 5 |
],
|
|
@@ -8,10 +8,10 @@
|
|
| 8 |
"AutoModel": "modeling_glm2.gLM2Model",
|
| 9 |
"AutoModelForMaskedLM": "modeling_glm2.gLM2ForMaskedLM"
|
| 10 |
},
|
| 11 |
-
"depth":
|
| 12 |
-
"dim":
|
| 13 |
"ffn_dim_multiplier": null,
|
| 14 |
-
"heads":
|
| 15 |
"model_type": "gLM2",
|
| 16 |
"norm_eps": 1e-05,
|
| 17 |
"swiglu_multiple_of": 256,
|
|
|
|
| 1 |
{
|
| 2 |
+
"_name_or_path": "tattabio/gLM2_150M",
|
| 3 |
"architectures": [
|
| 4 |
"gLM2ForMaskedLM"
|
| 5 |
],
|
|
|
|
| 8 |
"AutoModel": "modeling_glm2.gLM2Model",
|
| 9 |
"AutoModelForMaskedLM": "modeling_glm2.gLM2ForMaskedLM"
|
| 10 |
},
|
| 11 |
+
"depth": 30,
|
| 12 |
+
"dim": 640,
|
| 13 |
"ffn_dim_multiplier": null,
|
| 14 |
+
"heads": 10,
|
| 15 |
"model_type": "gLM2",
|
| 16 |
"norm_eps": 1e-05,
|
| 17 |
"swiglu_multiple_of": 256,
|
last-checkpoint/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d51c23a242cdd640054522d294ea271269733ca31b86fcbc629511f4b1ec40b4
|
| 3 |
+
size 609855088
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
-
size
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7fd4a9ab644e8a57e9ae9ade1220106ca660c1d95a05de03f2d5532224f2478b
|
| 3 |
+
size 1219840122
|
last-checkpoint/rng_state_0.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15024
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:848fdf35f13e1fde847fbd191021c99c0675e5e723a1b65fde4649f2fc9250db
|
| 3 |
size 15024
|
last-checkpoint/rng_state_1.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15024
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:d9f9fc41c0627d630837221d5c7872d3197c08985ee35f058d5f5e36bfe0249b
|
| 3 |
size 15024
|
last-checkpoint/rng_state_2.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15024
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7ae1449d711371210b0f6284f921f3df183a3c5c6628d3fc2950f5c89910866d
|
| 3 |
size 15024
|
last-checkpoint/rng_state_3.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 15024
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c3cc14e80a0475fa4dead8d6a3c6f0af9c5a92c40ad285584d68830834b3a6ea
|
| 3 |
size 15024
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1064
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:478ca537cf75a11344e25e46d3c46fdcf2db572bdb8cfff6f1ed3781e47a9787
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,572 +1,82 @@
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
-
"epoch": 0.
|
| 5 |
"eval_steps": 500,
|
| 6 |
-
"global_step":
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.0038360473368241363,
|
| 13 |
-
"grad_norm":
|
| 14 |
"learning_rate": 9.961638790854688e-05,
|
| 15 |
-
"loss":
|
| 16 |
"step": 50
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.0076720946736482725,
|
| 20 |
-
"grad_norm":
|
| 21 |
"learning_rate": 9.923277581709376e-05,
|
| 22 |
-
"loss":
|
| 23 |
"step": 100
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.011508142010472408,
|
| 27 |
-
"grad_norm":
|
| 28 |
"learning_rate": 9.884916372564063e-05,
|
| 29 |
-
"loss":
|
| 30 |
"step": 150
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.015344189347296545,
|
| 34 |
-
"grad_norm":
|
| 35 |
"learning_rate": 9.846555163418752e-05,
|
| 36 |
-
"loss":
|
| 37 |
"step": 200
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.019180236684120684,
|
| 41 |
-
"grad_norm":
|
| 42 |
"learning_rate": 9.80819395427344e-05,
|
| 43 |
-
"loss":
|
| 44 |
"step": 250
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.023016284020944817,
|
| 48 |
-
"grad_norm": 47.
|
| 49 |
"learning_rate": 9.769832745128127e-05,
|
| 50 |
-
"loss":
|
| 51 |
"step": 300
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.026852331357768953,
|
| 55 |
-
"grad_norm":
|
| 56 |
"learning_rate": 9.731471535982815e-05,
|
| 57 |
-
"loss":
|
| 58 |
"step": 350
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.03068837869459309,
|
| 62 |
-
"grad_norm":
|
| 63 |
"learning_rate": 9.693110326837502e-05,
|
| 64 |
-
"loss":
|
| 65 |
"step": 400
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.03452442603141723,
|
| 69 |
-
"grad_norm":
|
| 70 |
"learning_rate": 9.65474911769219e-05,
|
| 71 |
-
"loss":
|
| 72 |
"step": 450
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.03836047336824137,
|
| 76 |
-
"grad_norm":
|
| 77 |
"learning_rate": 9.616387908546877e-05,
|
| 78 |
-
"loss":
|
| 79 |
"step": 500
|
| 80 |
-
},
|
| 81 |
-
{
|
| 82 |
-
"epoch": 0.0421965207050655,
|
| 83 |
-
"grad_norm": 15.652584075927734,
|
| 84 |
-
"learning_rate": 9.578026699401565e-05,
|
| 85 |
-
"loss": 70.7517,
|
| 86 |
-
"step": 550
|
| 87 |
-
},
|
| 88 |
-
{
|
| 89 |
-
"epoch": 0.046032568041889634,
|
| 90 |
-
"grad_norm": 36.19481658935547,
|
| 91 |
-
"learning_rate": 9.539665490256254e-05,
|
| 92 |
-
"loss": 70.7868,
|
| 93 |
-
"step": 600
|
| 94 |
-
},
|
| 95 |
-
{
|
| 96 |
-
"epoch": 0.049868615378713774,
|
| 97 |
-
"grad_norm": 40.368431091308594,
|
| 98 |
-
"learning_rate": 9.501304281110941e-05,
|
| 99 |
-
"loss": 70.1634,
|
| 100 |
-
"step": 650
|
| 101 |
-
},
|
| 102 |
-
{
|
| 103 |
-
"epoch": 0.05370466271553791,
|
| 104 |
-
"grad_norm": 16.223190307617188,
|
| 105 |
-
"learning_rate": 9.462943071965629e-05,
|
| 106 |
-
"loss": 70.1049,
|
| 107 |
-
"step": 700
|
| 108 |
-
},
|
| 109 |
-
{
|
| 110 |
-
"epoch": 0.05754071005236205,
|
| 111 |
-
"grad_norm": 45.9012336730957,
|
| 112 |
-
"learning_rate": 9.424581862820317e-05,
|
| 113 |
-
"loss": 69.7258,
|
| 114 |
-
"step": 750
|
| 115 |
-
},
|
| 116 |
-
{
|
| 117 |
-
"epoch": 0.06137675738918618,
|
| 118 |
-
"grad_norm": 56.727088928222656,
|
| 119 |
-
"learning_rate": 9.386220653675004e-05,
|
| 120 |
-
"loss": 69.5,
|
| 121 |
-
"step": 800
|
| 122 |
-
},
|
| 123 |
-
{
|
| 124 |
-
"epoch": 0.06521280472601032,
|
| 125 |
-
"grad_norm": 41.01095199584961,
|
| 126 |
-
"learning_rate": 9.347859444529692e-05,
|
| 127 |
-
"loss": 69.2283,
|
| 128 |
-
"step": 850
|
| 129 |
-
},
|
| 130 |
-
{
|
| 131 |
-
"epoch": 0.06904885206283445,
|
| 132 |
-
"grad_norm": 49.74541091918945,
|
| 133 |
-
"learning_rate": 9.309498235384379e-05,
|
| 134 |
-
"loss": 69.0615,
|
| 135 |
-
"step": 900
|
| 136 |
-
},
|
| 137 |
-
{
|
| 138 |
-
"epoch": 0.07288489939965859,
|
| 139 |
-
"grad_norm": 24.164588928222656,
|
| 140 |
-
"learning_rate": 9.271137026239067e-05,
|
| 141 |
-
"loss": 68.7374,
|
| 142 |
-
"step": 950
|
| 143 |
-
},
|
| 144 |
-
{
|
| 145 |
-
"epoch": 0.07672094673648273,
|
| 146 |
-
"grad_norm": 44.137088775634766,
|
| 147 |
-
"learning_rate": 9.232775817093756e-05,
|
| 148 |
-
"loss": 68.4286,
|
| 149 |
-
"step": 1000
|
| 150 |
-
},
|
| 151 |
-
{
|
| 152 |
-
"epoch": 0.08055699407330687,
|
| 153 |
-
"grad_norm": 19.255796432495117,
|
| 154 |
-
"learning_rate": 9.194414607948443e-05,
|
| 155 |
-
"loss": 68.186,
|
| 156 |
-
"step": 1050
|
| 157 |
-
},
|
| 158 |
-
{
|
| 159 |
-
"epoch": 0.084393041410131,
|
| 160 |
-
"grad_norm": 44.74275207519531,
|
| 161 |
-
"learning_rate": 9.156053398803131e-05,
|
| 162 |
-
"loss": 68.2318,
|
| 163 |
-
"step": 1100
|
| 164 |
-
},
|
| 165 |
-
{
|
| 166 |
-
"epoch": 0.08822908874695513,
|
| 167 |
-
"grad_norm": 80.1006088256836,
|
| 168 |
-
"learning_rate": 9.117692189657818e-05,
|
| 169 |
-
"loss": 67.8507,
|
| 170 |
-
"step": 1150
|
| 171 |
-
},
|
| 172 |
-
{
|
| 173 |
-
"epoch": 0.09206513608377927,
|
| 174 |
-
"grad_norm": 34.77348327636719,
|
| 175 |
-
"learning_rate": 9.079330980512506e-05,
|
| 176 |
-
"loss": 67.8192,
|
| 177 |
-
"step": 1200
|
| 178 |
-
},
|
| 179 |
-
{
|
| 180 |
-
"epoch": 0.09590118342060341,
|
| 181 |
-
"grad_norm": 38.4120979309082,
|
| 182 |
-
"learning_rate": 9.040969771367193e-05,
|
| 183 |
-
"loss": 67.6707,
|
| 184 |
-
"step": 1250
|
| 185 |
-
},
|
| 186 |
-
{
|
| 187 |
-
"epoch": 0.09973723075742755,
|
| 188 |
-
"grad_norm": 19.280292510986328,
|
| 189 |
-
"learning_rate": 9.002608562221881e-05,
|
| 190 |
-
"loss": 67.4043,
|
| 191 |
-
"step": 1300
|
| 192 |
-
},
|
| 193 |
-
{
|
| 194 |
-
"epoch": 0.10357327809425168,
|
| 195 |
-
"grad_norm": 24.090499877929688,
|
| 196 |
-
"learning_rate": 8.964247353076569e-05,
|
| 197 |
-
"loss": 67.1605,
|
| 198 |
-
"step": 1350
|
| 199 |
-
},
|
| 200 |
-
{
|
| 201 |
-
"epoch": 0.10740932543107581,
|
| 202 |
-
"grad_norm": 40.173404693603516,
|
| 203 |
-
"learning_rate": 8.925886143931257e-05,
|
| 204 |
-
"loss": 66.9529,
|
| 205 |
-
"step": 1400
|
| 206 |
-
},
|
| 207 |
-
{
|
| 208 |
-
"epoch": 0.11124537276789996,
|
| 209 |
-
"grad_norm": 27.964691162109375,
|
| 210 |
-
"learning_rate": 8.887524934785945e-05,
|
| 211 |
-
"loss": 66.7643,
|
| 212 |
-
"step": 1450
|
| 213 |
-
},
|
| 214 |
-
{
|
| 215 |
-
"epoch": 0.1150814201047241,
|
| 216 |
-
"grad_norm": 20.457822799682617,
|
| 217 |
-
"learning_rate": 8.849163725640633e-05,
|
| 218 |
-
"loss": 66.7082,
|
| 219 |
-
"step": 1500
|
| 220 |
-
},
|
| 221 |
-
{
|
| 222 |
-
"epoch": 0.11891746744154823,
|
| 223 |
-
"grad_norm": 14.444588661193848,
|
| 224 |
-
"learning_rate": 8.81080251649532e-05,
|
| 225 |
-
"loss": 66.3239,
|
| 226 |
-
"step": 1550
|
| 227 |
-
},
|
| 228 |
-
{
|
| 229 |
-
"epoch": 0.12275351477837236,
|
| 230 |
-
"grad_norm": 45.22137451171875,
|
| 231 |
-
"learning_rate": 8.772441307350008e-05,
|
| 232 |
-
"loss": 66.273,
|
| 233 |
-
"step": 1600
|
| 234 |
-
},
|
| 235 |
-
{
|
| 236 |
-
"epoch": 0.1265895621151965,
|
| 237 |
-
"grad_norm": 23.911283493041992,
|
| 238 |
-
"learning_rate": 8.734080098204695e-05,
|
| 239 |
-
"loss": 65.9973,
|
| 240 |
-
"step": 1650
|
| 241 |
-
},
|
| 242 |
-
{
|
| 243 |
-
"epoch": 0.13042560945202064,
|
| 244 |
-
"grad_norm": 24.51435089111328,
|
| 245 |
-
"learning_rate": 8.695718889059383e-05,
|
| 246 |
-
"loss": 66.0171,
|
| 247 |
-
"step": 1700
|
| 248 |
-
},
|
| 249 |
-
{
|
| 250 |
-
"epoch": 0.13426165678884477,
|
| 251 |
-
"grad_norm": 24.943281173706055,
|
| 252 |
-
"learning_rate": 8.65735767991407e-05,
|
| 253 |
-
"loss": 65.7973,
|
| 254 |
-
"step": 1750
|
| 255 |
-
},
|
| 256 |
-
{
|
| 257 |
-
"epoch": 0.1380977041256689,
|
| 258 |
-
"grad_norm": 18.436256408691406,
|
| 259 |
-
"learning_rate": 8.618996470768759e-05,
|
| 260 |
-
"loss": 65.4202,
|
| 261 |
-
"step": 1800
|
| 262 |
-
},
|
| 263 |
-
{
|
| 264 |
-
"epoch": 0.14193375146249304,
|
| 265 |
-
"grad_norm": 36.64469909667969,
|
| 266 |
-
"learning_rate": 8.580635261623447e-05,
|
| 267 |
-
"loss": 65.2842,
|
| 268 |
-
"step": 1850
|
| 269 |
-
},
|
| 270 |
-
{
|
| 271 |
-
"epoch": 0.14576979879931717,
|
| 272 |
-
"grad_norm": 21.61631965637207,
|
| 273 |
-
"learning_rate": 8.542274052478134e-05,
|
| 274 |
-
"loss": 65.2321,
|
| 275 |
-
"step": 1900
|
| 276 |
-
},
|
| 277 |
-
{
|
| 278 |
-
"epoch": 0.1496058461361413,
|
| 279 |
-
"grad_norm": 26.61728286743164,
|
| 280 |
-
"learning_rate": 8.503912843332822e-05,
|
| 281 |
-
"loss": 65.193,
|
| 282 |
-
"step": 1950
|
| 283 |
-
},
|
| 284 |
-
{
|
| 285 |
-
"epoch": 0.15344189347296547,
|
| 286 |
-
"grad_norm": 41.06288528442383,
|
| 287 |
-
"learning_rate": 8.46555163418751e-05,
|
| 288 |
-
"loss": 64.8896,
|
| 289 |
-
"step": 2000
|
| 290 |
-
},
|
| 291 |
-
{
|
| 292 |
-
"epoch": 0.1572779408097896,
|
| 293 |
-
"grad_norm": 61.12138366699219,
|
| 294 |
-
"learning_rate": 8.427190425042197e-05,
|
| 295 |
-
"loss": 64.7165,
|
| 296 |
-
"step": 2050
|
| 297 |
-
},
|
| 298 |
-
{
|
| 299 |
-
"epoch": 0.16111398814661373,
|
| 300 |
-
"grad_norm": 43.435813903808594,
|
| 301 |
-
"learning_rate": 8.388829215896885e-05,
|
| 302 |
-
"loss": 64.6076,
|
| 303 |
-
"step": 2100
|
| 304 |
-
},
|
| 305 |
-
{
|
| 306 |
-
"epoch": 0.16495003548343787,
|
| 307 |
-
"grad_norm": 11.865663528442383,
|
| 308 |
-
"learning_rate": 8.350468006751573e-05,
|
| 309 |
-
"loss": 64.5384,
|
| 310 |
-
"step": 2150
|
| 311 |
-
},
|
| 312 |
-
{
|
| 313 |
-
"epoch": 0.168786082820262,
|
| 314 |
-
"grad_norm": 60.00083541870117,
|
| 315 |
-
"learning_rate": 8.312106797606261e-05,
|
| 316 |
-
"loss": 64.5437,
|
| 317 |
-
"step": 2200
|
| 318 |
-
},
|
| 319 |
-
{
|
| 320 |
-
"epoch": 0.17262213015708613,
|
| 321 |
-
"grad_norm": 37.82969284057617,
|
| 322 |
-
"learning_rate": 8.273745588460949e-05,
|
| 323 |
-
"loss": 64.0961,
|
| 324 |
-
"step": 2250
|
| 325 |
-
},
|
| 326 |
-
{
|
| 327 |
-
"epoch": 0.17645817749391027,
|
| 328 |
-
"grad_norm": 37.28142166137695,
|
| 329 |
-
"learning_rate": 8.235384379315636e-05,
|
| 330 |
-
"loss": 64.345,
|
| 331 |
-
"step": 2300
|
| 332 |
-
},
|
| 333 |
-
{
|
| 334 |
-
"epoch": 0.1802942248307344,
|
| 335 |
-
"grad_norm": 35.221309661865234,
|
| 336 |
-
"learning_rate": 8.197023170170324e-05,
|
| 337 |
-
"loss": 64.1263,
|
| 338 |
-
"step": 2350
|
| 339 |
-
},
|
| 340 |
-
{
|
| 341 |
-
"epoch": 0.18413027216755853,
|
| 342 |
-
"grad_norm": 22.304296493530273,
|
| 343 |
-
"learning_rate": 8.158661961025011e-05,
|
| 344 |
-
"loss": 63.8988,
|
| 345 |
-
"step": 2400
|
| 346 |
-
},
|
| 347 |
-
{
|
| 348 |
-
"epoch": 0.1879663195043827,
|
| 349 |
-
"grad_norm": 34.477996826171875,
|
| 350 |
-
"learning_rate": 8.120300751879699e-05,
|
| 351 |
-
"loss": 63.9893,
|
| 352 |
-
"step": 2450
|
| 353 |
-
},
|
| 354 |
-
{
|
| 355 |
-
"epoch": 0.19180236684120683,
|
| 356 |
-
"grad_norm": 85.30329895019531,
|
| 357 |
-
"learning_rate": 8.081939542734386e-05,
|
| 358 |
-
"loss": 63.6788,
|
| 359 |
-
"step": 2500
|
| 360 |
-
},
|
| 361 |
-
{
|
| 362 |
-
"epoch": 0.19563841417803096,
|
| 363 |
-
"grad_norm": 11.969497680664062,
|
| 364 |
-
"learning_rate": 8.043578333589075e-05,
|
| 365 |
-
"loss": 63.6427,
|
| 366 |
-
"step": 2550
|
| 367 |
-
},
|
| 368 |
-
{
|
| 369 |
-
"epoch": 0.1994744615148551,
|
| 370 |
-
"grad_norm": 37.35107421875,
|
| 371 |
-
"learning_rate": 8.005217124443763e-05,
|
| 372 |
-
"loss": 63.5714,
|
| 373 |
-
"step": 2600
|
| 374 |
-
},
|
| 375 |
-
{
|
| 376 |
-
"epoch": 0.20331050885167923,
|
| 377 |
-
"grad_norm": 44.66230773925781,
|
| 378 |
-
"learning_rate": 7.96685591529845e-05,
|
| 379 |
-
"loss": 63.2,
|
| 380 |
-
"step": 2650
|
| 381 |
-
},
|
| 382 |
-
{
|
| 383 |
-
"epoch": 0.20714655618850336,
|
| 384 |
-
"grad_norm": 53.52017593383789,
|
| 385 |
-
"learning_rate": 7.928494706153138e-05,
|
| 386 |
-
"loss": 63.0307,
|
| 387 |
-
"step": 2700
|
| 388 |
-
},
|
| 389 |
-
{
|
| 390 |
-
"epoch": 0.2109826035253275,
|
| 391 |
-
"grad_norm": 23.77840232849121,
|
| 392 |
-
"learning_rate": 7.890133497007825e-05,
|
| 393 |
-
"loss": 63.0493,
|
| 394 |
-
"step": 2750
|
| 395 |
-
},
|
| 396 |
-
{
|
| 397 |
-
"epoch": 0.21481865086215163,
|
| 398 |
-
"grad_norm": 52.308990478515625,
|
| 399 |
-
"learning_rate": 7.851772287862513e-05,
|
| 400 |
-
"loss": 62.8357,
|
| 401 |
-
"step": 2800
|
| 402 |
-
},
|
| 403 |
-
{
|
| 404 |
-
"epoch": 0.2186546981989758,
|
| 405 |
-
"grad_norm": 20.154300689697266,
|
| 406 |
-
"learning_rate": 7.8134110787172e-05,
|
| 407 |
-
"loss": 62.5052,
|
| 408 |
-
"step": 2850
|
| 409 |
-
},
|
| 410 |
-
{
|
| 411 |
-
"epoch": 0.22249074553579992,
|
| 412 |
-
"grad_norm": 15.035002708435059,
|
| 413 |
-
"learning_rate": 7.775049869571888e-05,
|
| 414 |
-
"loss": 62.5048,
|
| 415 |
-
"step": 2900
|
| 416 |
-
},
|
| 417 |
-
{
|
| 418 |
-
"epoch": 0.22632679287262406,
|
| 419 |
-
"grad_norm": 33.659698486328125,
|
| 420 |
-
"learning_rate": 7.736688660426577e-05,
|
| 421 |
-
"loss": 62.3453,
|
| 422 |
-
"step": 2950
|
| 423 |
-
},
|
| 424 |
-
{
|
| 425 |
-
"epoch": 0.2301628402094482,
|
| 426 |
-
"grad_norm": 55.17832946777344,
|
| 427 |
-
"learning_rate": 7.698327451281265e-05,
|
| 428 |
-
"loss": 62.4057,
|
| 429 |
-
"step": 3000
|
| 430 |
-
},
|
| 431 |
-
{
|
| 432 |
-
"epoch": 0.23399888754627232,
|
| 433 |
-
"grad_norm": 30.12982940673828,
|
| 434 |
-
"learning_rate": 7.659966242135954e-05,
|
| 435 |
-
"loss": 62.3078,
|
| 436 |
-
"step": 3050
|
| 437 |
-
},
|
| 438 |
-
{
|
| 439 |
-
"epoch": 0.23783493488309645,
|
| 440 |
-
"grad_norm": 42.589744567871094,
|
| 441 |
-
"learning_rate": 7.621605032990641e-05,
|
| 442 |
-
"loss": 62.3119,
|
| 443 |
-
"step": 3100
|
| 444 |
-
},
|
| 445 |
-
{
|
| 446 |
-
"epoch": 0.2416709822199206,
|
| 447 |
-
"grad_norm": 27.961658477783203,
|
| 448 |
-
"learning_rate": 7.583243823845329e-05,
|
| 449 |
-
"loss": 62.0758,
|
| 450 |
-
"step": 3150
|
| 451 |
-
},
|
| 452 |
-
{
|
| 453 |
-
"epoch": 0.24550702955674472,
|
| 454 |
-
"grad_norm": 24.691585540771484,
|
| 455 |
-
"learning_rate": 7.544882614700016e-05,
|
| 456 |
-
"loss": 61.9141,
|
| 457 |
-
"step": 3200
|
| 458 |
-
},
|
| 459 |
-
{
|
| 460 |
-
"epoch": 0.24934307689356885,
|
| 461 |
-
"grad_norm": 28.78058433532715,
|
| 462 |
-
"learning_rate": 7.506521405554704e-05,
|
| 463 |
-
"loss": 61.6389,
|
| 464 |
-
"step": 3250
|
| 465 |
-
},
|
| 466 |
-
{
|
| 467 |
-
"epoch": 0.253179124230393,
|
| 468 |
-
"grad_norm": 24.537670135498047,
|
| 469 |
-
"learning_rate": 7.468160196409391e-05,
|
| 470 |
-
"loss": 61.7158,
|
| 471 |
-
"step": 3300
|
| 472 |
-
},
|
| 473 |
-
{
|
| 474 |
-
"epoch": 0.25701517156721715,
|
| 475 |
-
"grad_norm": 29.42478370666504,
|
| 476 |
-
"learning_rate": 7.429798987264079e-05,
|
| 477 |
-
"loss": 61.9078,
|
| 478 |
-
"step": 3350
|
| 479 |
-
},
|
| 480 |
-
{
|
| 481 |
-
"epoch": 0.2608512189040413,
|
| 482 |
-
"grad_norm": 25.205659866333008,
|
| 483 |
-
"learning_rate": 7.391437778118768e-05,
|
| 484 |
-
"loss": 61.7029,
|
| 485 |
-
"step": 3400
|
| 486 |
-
},
|
| 487 |
-
{
|
| 488 |
-
"epoch": 0.2646872662408654,
|
| 489 |
-
"grad_norm": 37.85655212402344,
|
| 490 |
-
"learning_rate": 7.353076568973455e-05,
|
| 491 |
-
"loss": 61.5407,
|
| 492 |
-
"step": 3450
|
| 493 |
-
},
|
| 494 |
-
{
|
| 495 |
-
"epoch": 0.26852331357768955,
|
| 496 |
-
"grad_norm": 23.54515838623047,
|
| 497 |
-
"learning_rate": 7.314715359828143e-05,
|
| 498 |
-
"loss": 61.2952,
|
| 499 |
-
"step": 3500
|
| 500 |
-
},
|
| 501 |
-
{
|
| 502 |
-
"epoch": 0.2723593609145137,
|
| 503 |
-
"grad_norm": 22.69503402709961,
|
| 504 |
-
"learning_rate": 7.27635415068283e-05,
|
| 505 |
-
"loss": 61.2368,
|
| 506 |
-
"step": 3550
|
| 507 |
-
},
|
| 508 |
-
{
|
| 509 |
-
"epoch": 0.2761954082513378,
|
| 510 |
-
"grad_norm": 22.553903579711914,
|
| 511 |
-
"learning_rate": 7.237992941537518e-05,
|
| 512 |
-
"loss": 60.8539,
|
| 513 |
-
"step": 3600
|
| 514 |
-
},
|
| 515 |
-
{
|
| 516 |
-
"epoch": 0.28003145558816195,
|
| 517 |
-
"grad_norm": 56.599422454833984,
|
| 518 |
-
"learning_rate": 7.199631732392206e-05,
|
| 519 |
-
"loss": 60.8724,
|
| 520 |
-
"step": 3650
|
| 521 |
-
},
|
| 522 |
-
{
|
| 523 |
-
"epoch": 0.2838675029249861,
|
| 524 |
-
"grad_norm": 54.43523025512695,
|
| 525 |
-
"learning_rate": 7.161270523246893e-05,
|
| 526 |
-
"loss": 61.0066,
|
| 527 |
-
"step": 3700
|
| 528 |
-
},
|
| 529 |
-
{
|
| 530 |
-
"epoch": 0.2877035502618102,
|
| 531 |
-
"grad_norm": 21.187387466430664,
|
| 532 |
-
"learning_rate": 7.12290931410158e-05,
|
| 533 |
-
"loss": 60.7629,
|
| 534 |
-
"step": 3750
|
| 535 |
-
},
|
| 536 |
-
{
|
| 537 |
-
"epoch": 0.29153959759863435,
|
| 538 |
-
"grad_norm": 13.791070938110352,
|
| 539 |
-
"learning_rate": 7.08454810495627e-05,
|
| 540 |
-
"loss": 60.6954,
|
| 541 |
-
"step": 3800
|
| 542 |
-
},
|
| 543 |
-
{
|
| 544 |
-
"epoch": 0.2953756449354585,
|
| 545 |
-
"grad_norm": 18.809253692626953,
|
| 546 |
-
"learning_rate": 7.046186895810957e-05,
|
| 547 |
-
"loss": 60.6031,
|
| 548 |
-
"step": 3850
|
| 549 |
-
},
|
| 550 |
-
{
|
| 551 |
-
"epoch": 0.2992116922722826,
|
| 552 |
-
"grad_norm": 12.76346206665039,
|
| 553 |
-
"learning_rate": 7.007825686665645e-05,
|
| 554 |
-
"loss": 60.772,
|
| 555 |
-
"step": 3900
|
| 556 |
-
},
|
| 557 |
-
{
|
| 558 |
-
"epoch": 0.3030477396091068,
|
| 559 |
-
"grad_norm": 39.48878479003906,
|
| 560 |
-
"learning_rate": 6.969464477520332e-05,
|
| 561 |
-
"loss": 60.5169,
|
| 562 |
-
"step": 3950
|
| 563 |
-
},
|
| 564 |
-
{
|
| 565 |
-
"epoch": 0.30688378694593094,
|
| 566 |
-
"grad_norm": 61.519203186035156,
|
| 567 |
-
"learning_rate": 6.93110326837502e-05,
|
| 568 |
-
"loss": 60.3146,
|
| 569 |
-
"step": 4000
|
| 570 |
}
|
| 571 |
],
|
| 572 |
"logging_steps": 50,
|
|
@@ -586,7 +96,7 @@
|
|
| 586 |
"attributes": {}
|
| 587 |
}
|
| 588 |
},
|
| 589 |
-
"total_flos":
|
| 590 |
"train_batch_size": 2,
|
| 591 |
"trial_name": null,
|
| 592 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.03836047336824137,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 500,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
| 10 |
"log_history": [
|
| 11 |
{
|
| 12 |
"epoch": 0.0038360473368241363,
|
| 13 |
+
"grad_norm": 60.51473617553711,
|
| 14 |
"learning_rate": 9.961638790854688e-05,
|
| 15 |
+
"loss": 82.5331,
|
| 16 |
"step": 50
|
| 17 |
},
|
| 18 |
{
|
| 19 |
"epoch": 0.0076720946736482725,
|
| 20 |
+
"grad_norm": 41.30891418457031,
|
| 21 |
"learning_rate": 9.923277581709376e-05,
|
| 22 |
+
"loss": 79.5007,
|
| 23 |
"step": 100
|
| 24 |
},
|
| 25 |
{
|
| 26 |
"epoch": 0.011508142010472408,
|
| 27 |
+
"grad_norm": 47.033897399902344,
|
| 28 |
"learning_rate": 9.884916372564063e-05,
|
| 29 |
+
"loss": 78.5655,
|
| 30 |
"step": 150
|
| 31 |
},
|
| 32 |
{
|
| 33 |
"epoch": 0.015344189347296545,
|
| 34 |
+
"grad_norm": 47.13461685180664,
|
| 35 |
"learning_rate": 9.846555163418752e-05,
|
| 36 |
+
"loss": 77.9608,
|
| 37 |
"step": 200
|
| 38 |
},
|
| 39 |
{
|
| 40 |
"epoch": 0.019180236684120684,
|
| 41 |
+
"grad_norm": 50.00185012817383,
|
| 42 |
"learning_rate": 9.80819395427344e-05,
|
| 43 |
+
"loss": 77.3352,
|
| 44 |
"step": 250
|
| 45 |
},
|
| 46 |
{
|
| 47 |
"epoch": 0.023016284020944817,
|
| 48 |
+
"grad_norm": 47.55207443237305,
|
| 49 |
"learning_rate": 9.769832745128127e-05,
|
| 50 |
+
"loss": 76.8788,
|
| 51 |
"step": 300
|
| 52 |
},
|
| 53 |
{
|
| 54 |
"epoch": 0.026852331357768953,
|
| 55 |
+
"grad_norm": 32.375267028808594,
|
| 56 |
"learning_rate": 9.731471535982815e-05,
|
| 57 |
+
"loss": 76.3247,
|
| 58 |
"step": 350
|
| 59 |
},
|
| 60 |
{
|
| 61 |
"epoch": 0.03068837869459309,
|
| 62 |
+
"grad_norm": 26.361143112182617,
|
| 63 |
"learning_rate": 9.693110326837502e-05,
|
| 64 |
+
"loss": 76.1166,
|
| 65 |
"step": 400
|
| 66 |
},
|
| 67 |
{
|
| 68 |
"epoch": 0.03452442603141723,
|
| 69 |
+
"grad_norm": 30.172746658325195,
|
| 70 |
"learning_rate": 9.65474911769219e-05,
|
| 71 |
+
"loss": 75.7178,
|
| 72 |
"step": 450
|
| 73 |
},
|
| 74 |
{
|
| 75 |
"epoch": 0.03836047336824137,
|
| 76 |
+
"grad_norm": 48.87131118774414,
|
| 77 |
"learning_rate": 9.616387908546877e-05,
|
| 78 |
+
"loss": 75.2954,
|
| 79 |
"step": 500
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 80 |
}
|
| 81 |
],
|
| 82 |
"logging_steps": 50,
|
|
|
|
| 96 |
"attributes": {}
|
| 97 |
}
|
| 98 |
},
|
| 99 |
+
"total_flos": 1.9477233017238323e+17,
|
| 100 |
"train_batch_size": 2,
|
| 101 |
"trial_name": null,
|
| 102 |
"trial_params": null
|
last-checkpoint/training_args.bin
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 5240
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c1d0607b0ef32a7fb9356d7e9cea98c6b3f4ff9d058f42339926e1f0132b13a7
|
| 3 |
size 5240
|