Training in progress, step 1280, checkpoint
Browse files
last-checkpoint/model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 891644712
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:c19f4b8400a3385551bd98cf95b6e3f1d64dcf7f82712e395c277f316efdb089
|
| 3 |
size 891644712
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1783444794
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:84230522d2dcedf2de4d8922e543f648e253a14cc0db4e635e673f12800b1231
|
| 3 |
size 1783444794
|
last-checkpoint/rng_state.pth
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14244
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4e862bee55033739b9ae895cccb1fea0613d44a4ebc98463c3105553aed127ff
|
| 3 |
size 14244
|
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 sha256:9822aef898957bd458dc9360bb1e3058b7e31c090ed4f3ed492670e2394dfa96
|
| 3 |
size 1064
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -1,9 +1,9 @@
|
|
| 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,
|
|
@@ -4039,6 +4039,454 @@
|
|
| 4039 |
"learning_rate": 0.00015922098532995083,
|
| 4040 |
"loss": 0.5526,
|
| 4041 |
"step": 1152
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4042 |
}
|
| 4043 |
],
|
| 4044 |
"logging_steps": 2,
|
|
@@ -4058,7 +4506,7 @@
|
|
| 4058 |
"attributes": {}
|
| 4059 |
}
|
| 4060 |
},
|
| 4061 |
-
"total_flos":
|
| 4062 |
"train_batch_size": 8,
|
| 4063 |
"trial_name": null,
|
| 4064 |
"trial_params": null
|
|
|
|
| 1 |
{
|
| 2 |
"best_metric": null,
|
| 3 |
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 0.7099278979478647,
|
| 5 |
"eval_steps": 500,
|
| 6 |
+
"global_step": 1280,
|
| 7 |
"is_hyper_param_search": false,
|
| 8 |
"is_local_process_zero": true,
|
| 9 |
"is_world_process_zero": true,
|
|
|
|
| 4039 |
"learning_rate": 0.00015922098532995083,
|
| 4040 |
"loss": 0.5526,
|
| 4041 |
"step": 1152
|
| 4042 |
+
},
|
| 4043 |
+
{
|
| 4044 |
+
"epoch": 0.6400443704936217,
|
| 4045 |
+
"grad_norm": 0.2813306152820587,
|
| 4046 |
+
"learning_rate": 0.00015907611175727443,
|
| 4047 |
+
"loss": 0.4691,
|
| 4048 |
+
"step": 1154
|
| 4049 |
+
},
|
| 4050 |
+
{
|
| 4051 |
+
"epoch": 0.6411536328341653,
|
| 4052 |
+
"grad_norm": 0.27971866726875305,
|
| 4053 |
+
"learning_rate": 0.00015893104747161525,
|
| 4054 |
+
"loss": 0.4229,
|
| 4055 |
+
"step": 1156
|
| 4056 |
+
},
|
| 4057 |
+
{
|
| 4058 |
+
"epoch": 0.6422628951747088,
|
| 4059 |
+
"grad_norm": 0.2833665907382965,
|
| 4060 |
+
"learning_rate": 0.00015878579294127833,
|
| 4061 |
+
"loss": 0.3703,
|
| 4062 |
+
"step": 1158
|
| 4063 |
+
},
|
| 4064 |
+
{
|
| 4065 |
+
"epoch": 0.6433721575152523,
|
| 4066 |
+
"grad_norm": 0.2487824559211731,
|
| 4067 |
+
"learning_rate": 0.00015864034863518294,
|
| 4068 |
+
"loss": 0.4713,
|
| 4069 |
+
"step": 1160
|
| 4070 |
+
},
|
| 4071 |
+
{
|
| 4072 |
+
"epoch": 0.6444814198557959,
|
| 4073 |
+
"grad_norm": 0.2890799641609192,
|
| 4074 |
+
"learning_rate": 0.0001584947150228609,
|
| 4075 |
+
"loss": 0.4088,
|
| 4076 |
+
"step": 1162
|
| 4077 |
+
},
|
| 4078 |
+
{
|
| 4079 |
+
"epoch": 0.6455906821963394,
|
| 4080 |
+
"grad_norm": 0.38614341616630554,
|
| 4081 |
+
"learning_rate": 0.00015834889257445526,
|
| 4082 |
+
"loss": 0.6233,
|
| 4083 |
+
"step": 1164
|
| 4084 |
+
},
|
| 4085 |
+
{
|
| 4086 |
+
"epoch": 0.646699944536883,
|
| 4087 |
+
"grad_norm": 0.32349148392677307,
|
| 4088 |
+
"learning_rate": 0.00015820288176071861,
|
| 4089 |
+
"loss": 0.4304,
|
| 4090 |
+
"step": 1166
|
| 4091 |
+
},
|
| 4092 |
+
{
|
| 4093 |
+
"epoch": 0.6478092068774265,
|
| 4094 |
+
"grad_norm": 0.3689476549625397,
|
| 4095 |
+
"learning_rate": 0.0001580566830530117,
|
| 4096 |
+
"loss": 0.4791,
|
| 4097 |
+
"step": 1168
|
| 4098 |
+
},
|
| 4099 |
+
{
|
| 4100 |
+
"epoch": 0.64891846921797,
|
| 4101 |
+
"grad_norm": 0.33342450857162476,
|
| 4102 |
+
"learning_rate": 0.00015791029692330174,
|
| 4103 |
+
"loss": 0.5544,
|
| 4104 |
+
"step": 1170
|
| 4105 |
+
},
|
| 4106 |
+
{
|
| 4107 |
+
"epoch": 0.6500277315585136,
|
| 4108 |
+
"grad_norm": 0.27900078892707825,
|
| 4109 |
+
"learning_rate": 0.00015776372384416107,
|
| 4110 |
+
"loss": 0.3984,
|
| 4111 |
+
"step": 1172
|
| 4112 |
+
},
|
| 4113 |
+
{
|
| 4114 |
+
"epoch": 0.6511369938990571,
|
| 4115 |
+
"grad_norm": 0.2558038532733917,
|
| 4116 |
+
"learning_rate": 0.00015761696428876558,
|
| 4117 |
+
"loss": 0.3949,
|
| 4118 |
+
"step": 1174
|
| 4119 |
+
},
|
| 4120 |
+
{
|
| 4121 |
+
"epoch": 0.6522462562396006,
|
| 4122 |
+
"grad_norm": 0.3070181906223297,
|
| 4123 |
+
"learning_rate": 0.00015747001873089305,
|
| 4124 |
+
"loss": 0.4118,
|
| 4125 |
+
"step": 1176
|
| 4126 |
+
},
|
| 4127 |
+
{
|
| 4128 |
+
"epoch": 0.6533555185801442,
|
| 4129 |
+
"grad_norm": 0.3025922477245331,
|
| 4130 |
+
"learning_rate": 0.00015732288764492184,
|
| 4131 |
+
"loss": 0.4185,
|
| 4132 |
+
"step": 1178
|
| 4133 |
+
},
|
| 4134 |
+
{
|
| 4135 |
+
"epoch": 0.6544647809206877,
|
| 4136 |
+
"grad_norm": 0.2864340543746948,
|
| 4137 |
+
"learning_rate": 0.0001571755715058292,
|
| 4138 |
+
"loss": 0.4129,
|
| 4139 |
+
"step": 1180
|
| 4140 |
+
},
|
| 4141 |
+
{
|
| 4142 |
+
"epoch": 0.6555740432612313,
|
| 4143 |
+
"grad_norm": 0.30463624000549316,
|
| 4144 |
+
"learning_rate": 0.00015702807078918967,
|
| 4145 |
+
"loss": 0.3837,
|
| 4146 |
+
"step": 1182
|
| 4147 |
+
},
|
| 4148 |
+
{
|
| 4149 |
+
"epoch": 0.6566833056017748,
|
| 4150 |
+
"grad_norm": 0.36540600657463074,
|
| 4151 |
+
"learning_rate": 0.0001568803859711738,
|
| 4152 |
+
"loss": 0.4379,
|
| 4153 |
+
"step": 1184
|
| 4154 |
+
},
|
| 4155 |
+
{
|
| 4156 |
+
"epoch": 0.6577925679423183,
|
| 4157 |
+
"grad_norm": 0.30166247487068176,
|
| 4158 |
+
"learning_rate": 0.00015673251752854644,
|
| 4159 |
+
"loss": 0.4748,
|
| 4160 |
+
"step": 1186
|
| 4161 |
+
},
|
| 4162 |
+
{
|
| 4163 |
+
"epoch": 0.6589018302828619,
|
| 4164 |
+
"grad_norm": 0.2592626214027405,
|
| 4165 |
+
"learning_rate": 0.00015658446593866518,
|
| 4166 |
+
"loss": 0.4211,
|
| 4167 |
+
"step": 1188
|
| 4168 |
+
},
|
| 4169 |
+
{
|
| 4170 |
+
"epoch": 0.6600110926234054,
|
| 4171 |
+
"grad_norm": 0.3025457262992859,
|
| 4172 |
+
"learning_rate": 0.0001564362316794789,
|
| 4173 |
+
"loss": 0.4014,
|
| 4174 |
+
"step": 1190
|
| 4175 |
+
},
|
| 4176 |
+
{
|
| 4177 |
+
"epoch": 0.6611203549639489,
|
| 4178 |
+
"grad_norm": 0.3164603114128113,
|
| 4179 |
+
"learning_rate": 0.00015628781522952613,
|
| 4180 |
+
"loss": 0.3479,
|
| 4181 |
+
"step": 1192
|
| 4182 |
+
},
|
| 4183 |
+
{
|
| 4184 |
+
"epoch": 0.6622296173044925,
|
| 4185 |
+
"grad_norm": 0.3951297700405121,
|
| 4186 |
+
"learning_rate": 0.00015613921706793363,
|
| 4187 |
+
"loss": 0.5119,
|
| 4188 |
+
"step": 1194
|
| 4189 |
+
},
|
| 4190 |
+
{
|
| 4191 |
+
"epoch": 0.663338879645036,
|
| 4192 |
+
"grad_norm": 0.3717001974582672,
|
| 4193 |
+
"learning_rate": 0.00015599043767441473,
|
| 4194 |
+
"loss": 0.6534,
|
| 4195 |
+
"step": 1196
|
| 4196 |
+
},
|
| 4197 |
+
{
|
| 4198 |
+
"epoch": 0.6644481419855796,
|
| 4199 |
+
"grad_norm": 0.3157028555870056,
|
| 4200 |
+
"learning_rate": 0.0001558414775292678,
|
| 4201 |
+
"loss": 0.4616,
|
| 4202 |
+
"step": 1198
|
| 4203 |
+
},
|
| 4204 |
+
{
|
| 4205 |
+
"epoch": 0.6655574043261231,
|
| 4206 |
+
"grad_norm": 0.3201621472835541,
|
| 4207 |
+
"learning_rate": 0.00015569233711337476,
|
| 4208 |
+
"loss": 0.5525,
|
| 4209 |
+
"step": 1200
|
| 4210 |
+
},
|
| 4211 |
+
{
|
| 4212 |
+
"epoch": 0.6666666666666666,
|
| 4213 |
+
"grad_norm": 0.2802707254886627,
|
| 4214 |
+
"learning_rate": 0.00015554301690819952,
|
| 4215 |
+
"loss": 0.371,
|
| 4216 |
+
"step": 1202
|
| 4217 |
+
},
|
| 4218 |
+
{
|
| 4219 |
+
"epoch": 0.6677759290072102,
|
| 4220 |
+
"grad_norm": 0.2569611668586731,
|
| 4221 |
+
"learning_rate": 0.00015539351739578632,
|
| 4222 |
+
"loss": 0.4024,
|
| 4223 |
+
"step": 1204
|
| 4224 |
+
},
|
| 4225 |
+
{
|
| 4226 |
+
"epoch": 0.6688851913477537,
|
| 4227 |
+
"grad_norm": 0.20614203810691833,
|
| 4228 |
+
"learning_rate": 0.0001552438390587583,
|
| 4229 |
+
"loss": 0.4023,
|
| 4230 |
+
"step": 1206
|
| 4231 |
+
},
|
| 4232 |
+
{
|
| 4233 |
+
"epoch": 0.6699944536882972,
|
| 4234 |
+
"grad_norm": 0.28919216990470886,
|
| 4235 |
+
"learning_rate": 0.00015509398238031588,
|
| 4236 |
+
"loss": 0.5277,
|
| 4237 |
+
"step": 1208
|
| 4238 |
+
},
|
| 4239 |
+
{
|
| 4240 |
+
"epoch": 0.6711037160288408,
|
| 4241 |
+
"grad_norm": 0.3089144825935364,
|
| 4242 |
+
"learning_rate": 0.00015494394784423525,
|
| 4243 |
+
"loss": 0.4952,
|
| 4244 |
+
"step": 1210
|
| 4245 |
+
},
|
| 4246 |
+
{
|
| 4247 |
+
"epoch": 0.6722129783693843,
|
| 4248 |
+
"grad_norm": 0.3368174433708191,
|
| 4249 |
+
"learning_rate": 0.00015479373593486667,
|
| 4250 |
+
"loss": 0.5854,
|
| 4251 |
+
"step": 1212
|
| 4252 |
+
},
|
| 4253 |
+
{
|
| 4254 |
+
"epoch": 0.6733222407099279,
|
| 4255 |
+
"grad_norm": 0.45089298486709595,
|
| 4256 |
+
"learning_rate": 0.0001546433471371331,
|
| 4257 |
+
"loss": 0.4141,
|
| 4258 |
+
"step": 1214
|
| 4259 |
+
},
|
| 4260 |
+
{
|
| 4261 |
+
"epoch": 0.6744315030504714,
|
| 4262 |
+
"grad_norm": 0.3095158040523529,
|
| 4263 |
+
"learning_rate": 0.00015449278193652854,
|
| 4264 |
+
"loss": 0.4969,
|
| 4265 |
+
"step": 1216
|
| 4266 |
+
},
|
| 4267 |
+
{
|
| 4268 |
+
"epoch": 0.6755407653910149,
|
| 4269 |
+
"grad_norm": 0.36336401104927063,
|
| 4270 |
+
"learning_rate": 0.00015434204081911642,
|
| 4271 |
+
"loss": 0.4679,
|
| 4272 |
+
"step": 1218
|
| 4273 |
+
},
|
| 4274 |
+
{
|
| 4275 |
+
"epoch": 0.6766500277315585,
|
| 4276 |
+
"grad_norm": 0.3150896430015564,
|
| 4277 |
+
"learning_rate": 0.00015419112427152807,
|
| 4278 |
+
"loss": 0.5038,
|
| 4279 |
+
"step": 1220
|
| 4280 |
+
},
|
| 4281 |
+
{
|
| 4282 |
+
"epoch": 0.677759290072102,
|
| 4283 |
+
"grad_norm": 0.29561394453048706,
|
| 4284 |
+
"learning_rate": 0.0001540400327809612,
|
| 4285 |
+
"loss": 0.4697,
|
| 4286 |
+
"step": 1222
|
| 4287 |
+
},
|
| 4288 |
+
{
|
| 4289 |
+
"epoch": 0.6788685524126455,
|
| 4290 |
+
"grad_norm": 0.2975095212459564,
|
| 4291 |
+
"learning_rate": 0.00015388876683517826,
|
| 4292 |
+
"loss": 0.5544,
|
| 4293 |
+
"step": 1224
|
| 4294 |
+
},
|
| 4295 |
+
{
|
| 4296 |
+
"epoch": 0.6799778147531891,
|
| 4297 |
+
"grad_norm": 0.2124488651752472,
|
| 4298 |
+
"learning_rate": 0.00015373732692250486,
|
| 4299 |
+
"loss": 0.3321,
|
| 4300 |
+
"step": 1226
|
| 4301 |
+
},
|
| 4302 |
+
{
|
| 4303 |
+
"epoch": 0.6810870770937326,
|
| 4304 |
+
"grad_norm": 0.37978988885879517,
|
| 4305 |
+
"learning_rate": 0.00015358571353182824,
|
| 4306 |
+
"loss": 0.5268,
|
| 4307 |
+
"step": 1228
|
| 4308 |
+
},
|
| 4309 |
+
{
|
| 4310 |
+
"epoch": 0.6821963394342762,
|
| 4311 |
+
"grad_norm": 0.32258403301239014,
|
| 4312 |
+
"learning_rate": 0.0001534339271525957,
|
| 4313 |
+
"loss": 0.4983,
|
| 4314 |
+
"step": 1230
|
| 4315 |
+
},
|
| 4316 |
+
{
|
| 4317 |
+
"epoch": 0.6833056017748197,
|
| 4318 |
+
"grad_norm": 0.3182342052459717,
|
| 4319 |
+
"learning_rate": 0.00015328196827481302,
|
| 4320 |
+
"loss": 0.4742,
|
| 4321 |
+
"step": 1232
|
| 4322 |
+
},
|
| 4323 |
+
{
|
| 4324 |
+
"epoch": 0.6844148641153632,
|
| 4325 |
+
"grad_norm": 0.26553046703338623,
|
| 4326 |
+
"learning_rate": 0.0001531298373890427,
|
| 4327 |
+
"loss": 0.4627,
|
| 4328 |
+
"step": 1234
|
| 4329 |
+
},
|
| 4330 |
+
{
|
| 4331 |
+
"epoch": 0.6855241264559068,
|
| 4332 |
+
"grad_norm": 0.3853413164615631,
|
| 4333 |
+
"learning_rate": 0.0001529775349864026,
|
| 4334 |
+
"loss": 0.5493,
|
| 4335 |
+
"step": 1236
|
| 4336 |
+
},
|
| 4337 |
+
{
|
| 4338 |
+
"epoch": 0.6866333887964503,
|
| 4339 |
+
"grad_norm": 0.26181191205978394,
|
| 4340 |
+
"learning_rate": 0.0001528250615585644,
|
| 4341 |
+
"loss": 0.4308,
|
| 4342 |
+
"step": 1238
|
| 4343 |
+
},
|
| 4344 |
+
{
|
| 4345 |
+
"epoch": 0.687742651136994,
|
| 4346 |
+
"grad_norm": 0.29632505774497986,
|
| 4347 |
+
"learning_rate": 0.0001526724175977518,
|
| 4348 |
+
"loss": 0.3972,
|
| 4349 |
+
"step": 1240
|
| 4350 |
+
},
|
| 4351 |
+
{
|
| 4352 |
+
"epoch": 0.6888519134775375,
|
| 4353 |
+
"grad_norm": 0.21180076897144318,
|
| 4354 |
+
"learning_rate": 0.000152519603596739,
|
| 4355 |
+
"loss": 0.2871,
|
| 4356 |
+
"step": 1242
|
| 4357 |
+
},
|
| 4358 |
+
{
|
| 4359 |
+
"epoch": 0.689961175818081,
|
| 4360 |
+
"grad_norm": 0.3028866946697235,
|
| 4361 |
+
"learning_rate": 0.00015236662004884912,
|
| 4362 |
+
"loss": 0.5045,
|
| 4363 |
+
"step": 1244
|
| 4364 |
+
},
|
| 4365 |
+
{
|
| 4366 |
+
"epoch": 0.6910704381586246,
|
| 4367 |
+
"grad_norm": 0.22230632603168488,
|
| 4368 |
+
"learning_rate": 0.0001522134674479527,
|
| 4369 |
+
"loss": 0.3913,
|
| 4370 |
+
"step": 1246
|
| 4371 |
+
},
|
| 4372 |
+
{
|
| 4373 |
+
"epoch": 0.6921797004991681,
|
| 4374 |
+
"grad_norm": 0.3115563690662384,
|
| 4375 |
+
"learning_rate": 0.00015206014628846594,
|
| 4376 |
+
"loss": 0.4612,
|
| 4377 |
+
"step": 1248
|
| 4378 |
+
},
|
| 4379 |
+
{
|
| 4380 |
+
"epoch": 0.6932889628397116,
|
| 4381 |
+
"grad_norm": 0.30068281292915344,
|
| 4382 |
+
"learning_rate": 0.00015190665706534925,
|
| 4383 |
+
"loss": 0.4224,
|
| 4384 |
+
"step": 1250
|
| 4385 |
+
},
|
| 4386 |
+
{
|
| 4387 |
+
"epoch": 0.6943982251802552,
|
| 4388 |
+
"grad_norm": 0.3180410861968994,
|
| 4389 |
+
"learning_rate": 0.00015175300027410566,
|
| 4390 |
+
"loss": 0.4094,
|
| 4391 |
+
"step": 1252
|
| 4392 |
+
},
|
| 4393 |
+
{
|
| 4394 |
+
"epoch": 0.6955074875207987,
|
| 4395 |
+
"grad_norm": 0.3450130224227905,
|
| 4396 |
+
"learning_rate": 0.00015159917641077895,
|
| 4397 |
+
"loss": 0.5635,
|
| 4398 |
+
"step": 1254
|
| 4399 |
+
},
|
| 4400 |
+
{
|
| 4401 |
+
"epoch": 0.6966167498613423,
|
| 4402 |
+
"grad_norm": 0.27590128779411316,
|
| 4403 |
+
"learning_rate": 0.00015144518597195243,
|
| 4404 |
+
"loss": 0.4893,
|
| 4405 |
+
"step": 1256
|
| 4406 |
+
},
|
| 4407 |
+
{
|
| 4408 |
+
"epoch": 0.6977260122018858,
|
| 4409 |
+
"grad_norm": 0.25257548689842224,
|
| 4410 |
+
"learning_rate": 0.0001512910294547471,
|
| 4411 |
+
"loss": 0.4252,
|
| 4412 |
+
"step": 1258
|
| 4413 |
+
},
|
| 4414 |
+
{
|
| 4415 |
+
"epoch": 0.6988352745424293,
|
| 4416 |
+
"grad_norm": 0.3455219864845276,
|
| 4417 |
+
"learning_rate": 0.00015113670735682013,
|
| 4418 |
+
"loss": 0.5274,
|
| 4419 |
+
"step": 1260
|
| 4420 |
+
},
|
| 4421 |
+
{
|
| 4422 |
+
"epoch": 0.6999445368829729,
|
| 4423 |
+
"grad_norm": 0.3079371750354767,
|
| 4424 |
+
"learning_rate": 0.0001509822201763632,
|
| 4425 |
+
"loss": 0.3667,
|
| 4426 |
+
"step": 1262
|
| 4427 |
+
},
|
| 4428 |
+
{
|
| 4429 |
+
"epoch": 0.7010537992235164,
|
| 4430 |
+
"grad_norm": 0.27760395407676697,
|
| 4431 |
+
"learning_rate": 0.00015082756841210086,
|
| 4432 |
+
"loss": 0.4693,
|
| 4433 |
+
"step": 1264
|
| 4434 |
+
},
|
| 4435 |
+
{
|
| 4436 |
+
"epoch": 0.7021630615640599,
|
| 4437 |
+
"grad_norm": 0.3691544830799103,
|
| 4438 |
+
"learning_rate": 0.0001506727525632891,
|
| 4439 |
+
"loss": 0.4975,
|
| 4440 |
+
"step": 1266
|
| 4441 |
+
},
|
| 4442 |
+
{
|
| 4443 |
+
"epoch": 0.7032723239046035,
|
| 4444 |
+
"grad_norm": 0.30312782526016235,
|
| 4445 |
+
"learning_rate": 0.00015051777312971357,
|
| 4446 |
+
"loss": 0.4377,
|
| 4447 |
+
"step": 1268
|
| 4448 |
+
},
|
| 4449 |
+
{
|
| 4450 |
+
"epoch": 0.704381586245147,
|
| 4451 |
+
"grad_norm": 0.31264883279800415,
|
| 4452 |
+
"learning_rate": 0.00015036263061168797,
|
| 4453 |
+
"loss": 0.3841,
|
| 4454 |
+
"step": 1270
|
| 4455 |
+
},
|
| 4456 |
+
{
|
| 4457 |
+
"epoch": 0.7054908485856906,
|
| 4458 |
+
"grad_norm": 0.27693864703178406,
|
| 4459 |
+
"learning_rate": 0.0001502073255100525,
|
| 4460 |
+
"loss": 0.4347,
|
| 4461 |
+
"step": 1272
|
| 4462 |
+
},
|
| 4463 |
+
{
|
| 4464 |
+
"epoch": 0.7066001109262341,
|
| 4465 |
+
"grad_norm": 0.3322737514972687,
|
| 4466 |
+
"learning_rate": 0.0001500518583261723,
|
| 4467 |
+
"loss": 0.4424,
|
| 4468 |
+
"step": 1274
|
| 4469 |
+
},
|
| 4470 |
+
{
|
| 4471 |
+
"epoch": 0.7077093732667776,
|
| 4472 |
+
"grad_norm": 0.2735716998577118,
|
| 4473 |
+
"learning_rate": 0.0001498962295619356,
|
| 4474 |
+
"loss": 0.3577,
|
| 4475 |
+
"step": 1276
|
| 4476 |
+
},
|
| 4477 |
+
{
|
| 4478 |
+
"epoch": 0.7088186356073212,
|
| 4479 |
+
"grad_norm": 0.2565724849700928,
|
| 4480 |
+
"learning_rate": 0.00014974043971975243,
|
| 4481 |
+
"loss": 0.4086,
|
| 4482 |
+
"step": 1278
|
| 4483 |
+
},
|
| 4484 |
+
{
|
| 4485 |
+
"epoch": 0.7099278979478647,
|
| 4486 |
+
"grad_norm": 0.4010816216468811,
|
| 4487 |
+
"learning_rate": 0.00014958448930255265,
|
| 4488 |
+
"loss": 0.5353,
|
| 4489 |
+
"step": 1280
|
| 4490 |
}
|
| 4491 |
],
|
| 4492 |
"logging_steps": 2,
|
|
|
|
| 4506 |
"attributes": {}
|
| 4507 |
}
|
| 4508 |
},
|
| 4509 |
+
"total_flos": 3117864399667200.0,
|
| 4510 |
"train_batch_size": 8,
|
| 4511 |
"trial_name": null,
|
| 4512 |
"trial_params": null
|