Training in progress, step 4200
Browse files- adapter_model.safetensors +1 -1
- last-checkpoint/adapter_config.json +3 -3
- last-checkpoint/adapter_model.safetensors +1 -1
- last-checkpoint/optimizer.pt +1 -1
- last-checkpoint/rng_state_0.pth +1 -1
- last-checkpoint/rng_state_1.pth +1 -1
- last-checkpoint/scaler.pt +1 -1
- last-checkpoint/scheduler.pt +1 -1
- last-checkpoint/trainer_state.json +3 -447
adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14764744
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e82bf428995d67e4e58c19e3e3f3979adfcb76ca6133f8e4c6eb8d01eba3a991
|
| 3 |
size 14764744
|
last-checkpoint/adapter_config.json
CHANGED
|
@@ -31,10 +31,10 @@
|
|
| 31 |
"rank_pattern": {},
|
| 32 |
"revision": null,
|
| 33 |
"target_modules": [
|
| 34 |
-
"q_proj",
|
| 35 |
"v_proj",
|
| 36 |
-
"
|
| 37 |
-
"
|
|
|
|
| 38 |
],
|
| 39 |
"target_parameters": null,
|
| 40 |
"task_type": "CAUSAL_LM",
|
|
|
|
| 31 |
"rank_pattern": {},
|
| 32 |
"revision": null,
|
| 33 |
"target_modules": [
|
|
|
|
| 34 |
"v_proj",
|
| 35 |
+
"k_proj",
|
| 36 |
+
"q_proj",
|
| 37 |
+
"o_proj"
|
| 38 |
],
|
| 39 |
"target_parameters": null,
|
| 40 |
"task_type": "CAUSAL_LM",
|
last-checkpoint/adapter_model.safetensors
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 14764744
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:607b476259c63c4d83dfb124f527cefa79bd00fe22b5a14b7b344dafd9bf5ea1
|
| 3 |
size 14764744
|
last-checkpoint/optimizer.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 29614667
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:0282f8c51c898347bdf2ebe27e230c3c8341fbe228246855db3fe9ce0a66fa7d
|
| 3 |
size 29614667
|
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 14917
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:88f174f1066ea4a4da640e826a26e458fb763486cce7102ce575770a7049aa0b
|
| 3 |
size 14917
|
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 14917
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:ee6c99225d51e4516a7663ba39f71bc213571f59ec57cba08f558316c085c20f
|
| 3 |
size 14917
|
last-checkpoint/scaler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1383
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:7788c7cc60bd991bb584c9a8631c7010707cb818c6f2829b9625a72fb3ec1a89
|
| 3 |
size 1383
|
last-checkpoint/scheduler.pt
CHANGED
|
@@ -1,3 +1,3 @@
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
-
oid sha256:
|
| 3 |
size 1465
|
|
|
|
| 1 |
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a610fcd2a240821c3e97e46e44aaefc0d8bb3f7e501d702e1f72a0ed2d71278b
|
| 3 |
size 1465
|
last-checkpoint/trainer_state.json
CHANGED
|
@@ -2,9 +2,9 @@
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
-
"epoch": 0.
|
| 6 |
"eval_steps": 100,
|
| 7 |
-
"global_step":
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
@@ -4004,450 +4004,6 @@
|
|
| 4004 |
"eval_samples_per_second": 5.088,
|
| 4005 |
"eval_steps_per_second": 1.272,
|
| 4006 |
"step": 3600
|
| 4007 |
-
},
|
| 4008 |
-
{
|
| 4009 |
-
"entropy": 0.8791561007499695,
|
| 4010 |
-
"epoch": 0.621797356069414,
|
| 4011 |
-
"grad_norm": 0.4086858332157135,
|
| 4012 |
-
"learning_rate": 7.568033069238719e-06,
|
| 4013 |
-
"loss": 0.8722482681274414,
|
| 4014 |
-
"mean_token_accuracy": 0.8025152161717415,
|
| 4015 |
-
"num_tokens": 132977.0,
|
| 4016 |
-
"step": 3610
|
| 4017 |
-
},
|
| 4018 |
-
{
|
| 4019 |
-
"entropy": 0.899948114156723,
|
| 4020 |
-
"epoch": 0.6235197864186367,
|
| 4021 |
-
"grad_norm": 0.46601250767707825,
|
| 4022 |
-
"learning_rate": 7.533585945573545e-06,
|
| 4023 |
-
"loss": 0.8855493545532227,
|
| 4024 |
-
"mean_token_accuracy": 0.800511172413826,
|
| 4025 |
-
"num_tokens": 272879.0,
|
| 4026 |
-
"step": 3620
|
| 4027 |
-
},
|
| 4028 |
-
{
|
| 4029 |
-
"entropy": 0.8673037841916085,
|
| 4030 |
-
"epoch": 0.6252422167678594,
|
| 4031 |
-
"grad_norm": 0.4145843982696533,
|
| 4032 |
-
"learning_rate": 7.499138821908372e-06,
|
| 4033 |
-
"loss": 0.8640913963317871,
|
| 4034 |
-
"mean_token_accuracy": 0.803620132803917,
|
| 4035 |
-
"num_tokens": 410749.0,
|
| 4036 |
-
"step": 3630
|
| 4037 |
-
},
|
| 4038 |
-
{
|
| 4039 |
-
"entropy": 0.8860716938972473,
|
| 4040 |
-
"epoch": 0.6269646471170822,
|
| 4041 |
-
"grad_norm": 0.42249223589897156,
|
| 4042 |
-
"learning_rate": 7.464691698243198e-06,
|
| 4043 |
-
"loss": 0.8842039108276367,
|
| 4044 |
-
"mean_token_accuracy": 0.7986636623740196,
|
| 4045 |
-
"num_tokens": 547722.0,
|
| 4046 |
-
"step": 3640
|
| 4047 |
-
},
|
| 4048 |
-
{
|
| 4049 |
-
"entropy": 0.8954877465963363,
|
| 4050 |
-
"epoch": 0.6286870774663049,
|
| 4051 |
-
"grad_norm": 0.3925943672657013,
|
| 4052 |
-
"learning_rate": 7.4302445745780235e-06,
|
| 4053 |
-
"loss": 0.8855221748352051,
|
| 4054 |
-
"mean_token_accuracy": 0.7995046123862266,
|
| 4055 |
-
"num_tokens": 684572.0,
|
| 4056 |
-
"step": 3650
|
| 4057 |
-
},
|
| 4058 |
-
{
|
| 4059 |
-
"entropy": 0.8849117711186409,
|
| 4060 |
-
"epoch": 0.6304095078155277,
|
| 4061 |
-
"grad_norm": 0.4629289507865906,
|
| 4062 |
-
"learning_rate": 7.395797450912849e-06,
|
| 4063 |
-
"loss": 0.8729182243347168,
|
| 4064 |
-
"mean_token_accuracy": 0.8043904855847359,
|
| 4065 |
-
"num_tokens": 827644.0,
|
| 4066 |
-
"step": 3660
|
| 4067 |
-
},
|
| 4068 |
-
{
|
| 4069 |
-
"entropy": 0.8804656386375427,
|
| 4070 |
-
"epoch": 0.6321319381647504,
|
| 4071 |
-
"grad_norm": 0.46771207451820374,
|
| 4072 |
-
"learning_rate": 7.361350327247675e-06,
|
| 4073 |
-
"loss": 0.8755840301513672,
|
| 4074 |
-
"mean_token_accuracy": 0.8031970843672752,
|
| 4075 |
-
"num_tokens": 962324.0,
|
| 4076 |
-
"step": 3670
|
| 4077 |
-
},
|
| 4078 |
-
{
|
| 4079 |
-
"entropy": 0.9434955894947052,
|
| 4080 |
-
"epoch": 0.6338543685139733,
|
| 4081 |
-
"grad_norm": 0.5211445689201355,
|
| 4082 |
-
"learning_rate": 7.326903203582502e-06,
|
| 4083 |
-
"loss": 0.9384965896606445,
|
| 4084 |
-
"mean_token_accuracy": 0.7892407014966011,
|
| 4085 |
-
"num_tokens": 1098423.0,
|
| 4086 |
-
"step": 3680
|
| 4087 |
-
},
|
| 4088 |
-
{
|
| 4089 |
-
"entropy": 0.8946269884705543,
|
| 4090 |
-
"epoch": 0.635576798863196,
|
| 4091 |
-
"grad_norm": 0.5688809752464294,
|
| 4092 |
-
"learning_rate": 7.292456079917327e-06,
|
| 4093 |
-
"loss": 0.8830915451049804,
|
| 4094 |
-
"mean_token_accuracy": 0.8004635199904442,
|
| 4095 |
-
"num_tokens": 1236560.0,
|
| 4096 |
-
"step": 3690
|
| 4097 |
-
},
|
| 4098 |
-
{
|
| 4099 |
-
"entropy": 0.8900304272770881,
|
| 4100 |
-
"epoch": 0.6372992292124188,
|
| 4101 |
-
"grad_norm": 0.38199931383132935,
|
| 4102 |
-
"learning_rate": 7.258008956252154e-06,
|
| 4103 |
-
"loss": 0.8838729858398438,
|
| 4104 |
-
"mean_token_accuracy": 0.8015081480145454,
|
| 4105 |
-
"num_tokens": 1374095.0,
|
| 4106 |
-
"step": 3700
|
| 4107 |
-
},
|
| 4108 |
-
{
|
| 4109 |
-
"epoch": 0.6372992292124188,
|
| 4110 |
-
"eval_entropy": 0.8987031751527014,
|
| 4111 |
-
"eval_loss": 0.8895877599716187,
|
| 4112 |
-
"eval_mean_token_accuracy": 0.7999683152285254,
|
| 4113 |
-
"eval_num_tokens": 1374095.0,
|
| 4114 |
-
"eval_runtime": 1996.9933,
|
| 4115 |
-
"eval_samples_per_second": 5.168,
|
| 4116 |
-
"eval_steps_per_second": 1.292,
|
| 4117 |
-
"step": 3700
|
| 4118 |
-
},
|
| 4119 |
-
{
|
| 4120 |
-
"entropy": 0.8338712751865387,
|
| 4121 |
-
"epoch": 0.6390216595616415,
|
| 4122 |
-
"grad_norm": 0.38810011744499207,
|
| 4123 |
-
"learning_rate": 7.22356183258698e-06,
|
| 4124 |
-
"loss": 0.8263821601867676,
|
| 4125 |
-
"mean_token_accuracy": 0.8129131481051445,
|
| 4126 |
-
"num_tokens": 1513665.0,
|
| 4127 |
-
"step": 3710
|
| 4128 |
-
},
|
| 4129 |
-
{
|
| 4130 |
-
"entropy": 0.8491598337888717,
|
| 4131 |
-
"epoch": 0.6407440899108642,
|
| 4132 |
-
"grad_norm": 0.5338125228881836,
|
| 4133 |
-
"learning_rate": 7.189114708921806e-06,
|
| 4134 |
-
"loss": 0.8426370620727539,
|
| 4135 |
-
"mean_token_accuracy": 0.8097073540091515,
|
| 4136 |
-
"num_tokens": 1655522.0,
|
| 4137 |
-
"step": 3720
|
| 4138 |
-
},
|
| 4139 |
-
{
|
| 4140 |
-
"entropy": 0.8653292268514633,
|
| 4141 |
-
"epoch": 0.642466520260087,
|
| 4142 |
-
"grad_norm": 0.4272521734237671,
|
| 4143 |
-
"learning_rate": 7.154667585256631e-06,
|
| 4144 |
-
"loss": 0.8501546859741211,
|
| 4145 |
-
"mean_token_accuracy": 0.8075862035155297,
|
| 4146 |
-
"num_tokens": 1792787.0,
|
| 4147 |
-
"step": 3730
|
| 4148 |
-
},
|
| 4149 |
-
{
|
| 4150 |
-
"entropy": 0.86524413600564,
|
| 4151 |
-
"epoch": 0.6441889506093097,
|
| 4152 |
-
"grad_norm": 0.40841034054756165,
|
| 4153 |
-
"learning_rate": 7.120220461591457e-06,
|
| 4154 |
-
"loss": 0.8654015541076661,
|
| 4155 |
-
"mean_token_accuracy": 0.8041766241192818,
|
| 4156 |
-
"num_tokens": 1931208.0,
|
| 4157 |
-
"step": 3740
|
| 4158 |
-
},
|
| 4159 |
-
{
|
| 4160 |
-
"entropy": 0.8816385686397552,
|
| 4161 |
-
"epoch": 0.6459113809585325,
|
| 4162 |
-
"grad_norm": 0.4200495183467865,
|
| 4163 |
-
"learning_rate": 7.085773337926284e-06,
|
| 4164 |
-
"loss": 0.8681506156921387,
|
| 4165 |
-
"mean_token_accuracy": 0.8023644983768463,
|
| 4166 |
-
"num_tokens": 2069218.0,
|
| 4167 |
-
"step": 3750
|
| 4168 |
-
},
|
| 4169 |
-
{
|
| 4170 |
-
"entropy": 0.8531557321548462,
|
| 4171 |
-
"epoch": 0.6476338113077552,
|
| 4172 |
-
"grad_norm": 0.4340897500514984,
|
| 4173 |
-
"learning_rate": 7.051326214261109e-06,
|
| 4174 |
-
"loss": 0.8407665252685547,
|
| 4175 |
-
"mean_token_accuracy": 0.8092893213033676,
|
| 4176 |
-
"num_tokens": 2207220.0,
|
| 4177 |
-
"step": 3760
|
| 4178 |
-
},
|
| 4179 |
-
{
|
| 4180 |
-
"entropy": 0.9071142002940178,
|
| 4181 |
-
"epoch": 0.649356241656978,
|
| 4182 |
-
"grad_norm": 0.4704848527908325,
|
| 4183 |
-
"learning_rate": 7.016879090595936e-06,
|
| 4184 |
-
"loss": 0.9017156600952149,
|
| 4185 |
-
"mean_token_accuracy": 0.7971421957015992,
|
| 4186 |
-
"num_tokens": 2341438.0,
|
| 4187 |
-
"step": 3770
|
| 4188 |
-
},
|
| 4189 |
-
{
|
| 4190 |
-
"entropy": 0.9055481895804405,
|
| 4191 |
-
"epoch": 0.6510786720062007,
|
| 4192 |
-
"grad_norm": 0.39707693457603455,
|
| 4193 |
-
"learning_rate": 6.982431966930762e-06,
|
| 4194 |
-
"loss": 0.9052577972412109,
|
| 4195 |
-
"mean_token_accuracy": 0.7952045202255249,
|
| 4196 |
-
"num_tokens": 2482296.0,
|
| 4197 |
-
"step": 3780
|
| 4198 |
-
},
|
| 4199 |
-
{
|
| 4200 |
-
"entropy": 0.8833331823348999,
|
| 4201 |
-
"epoch": 0.6528011023554235,
|
| 4202 |
-
"grad_norm": 0.416567862033844,
|
| 4203 |
-
"learning_rate": 6.9479848432655884e-06,
|
| 4204 |
-
"loss": 0.8772726058959961,
|
| 4205 |
-
"mean_token_accuracy": 0.8013244524598122,
|
| 4206 |
-
"num_tokens": 2622763.0,
|
| 4207 |
-
"step": 3790
|
| 4208 |
-
},
|
| 4209 |
-
{
|
| 4210 |
-
"entropy": 0.8962781682610512,
|
| 4211 |
-
"epoch": 0.6545235327046462,
|
| 4212 |
-
"grad_norm": 0.41791078448295593,
|
| 4213 |
-
"learning_rate": 6.9135377196004135e-06,
|
| 4214 |
-
"loss": 0.8844639778137207,
|
| 4215 |
-
"mean_token_accuracy": 0.8016847550868988,
|
| 4216 |
-
"num_tokens": 2756826.0,
|
| 4217 |
-
"step": 3800
|
| 4218 |
-
},
|
| 4219 |
-
{
|
| 4220 |
-
"epoch": 0.6545235327046462,
|
| 4221 |
-
"eval_entropy": 0.8984109290876577,
|
| 4222 |
-
"eval_loss": 0.8893030881881714,
|
| 4223 |
-
"eval_mean_token_accuracy": 0.8000404531297088,
|
| 4224 |
-
"eval_num_tokens": 2756826.0,
|
| 4225 |
-
"eval_runtime": 1997.5778,
|
| 4226 |
-
"eval_samples_per_second": 5.167,
|
| 4227 |
-
"eval_steps_per_second": 1.292,
|
| 4228 |
-
"step": 3800
|
| 4229 |
-
},
|
| 4230 |
-
{
|
| 4231 |
-
"entropy": 0.8708895847201348,
|
| 4232 |
-
"epoch": 0.6562459630538691,
|
| 4233 |
-
"grad_norm": 0.3722359538078308,
|
| 4234 |
-
"learning_rate": 6.879090595935239e-06,
|
| 4235 |
-
"loss": 0.8560014724731445,
|
| 4236 |
-
"mean_token_accuracy": 0.8055260598659515,
|
| 4237 |
-
"num_tokens": 2896570.0,
|
| 4238 |
-
"step": 3810
|
| 4239 |
-
},
|
| 4240 |
-
{
|
| 4241 |
-
"entropy": 0.8795327812433242,
|
| 4242 |
-
"epoch": 0.6579683934030918,
|
| 4243 |
-
"grad_norm": 0.4038887917995453,
|
| 4244 |
-
"learning_rate": 6.844643472270066e-06,
|
| 4245 |
-
"loss": 0.8748892784118653,
|
| 4246 |
-
"mean_token_accuracy": 0.8010203137993812,
|
| 4247 |
-
"num_tokens": 3036528.0,
|
| 4248 |
-
"step": 3820
|
| 4249 |
-
},
|
| 4250 |
-
{
|
| 4251 |
-
"entropy": 0.8723071977496147,
|
| 4252 |
-
"epoch": 0.6596908237523145,
|
| 4253 |
-
"grad_norm": 0.38578343391418457,
|
| 4254 |
-
"learning_rate": 6.810196348604892e-06,
|
| 4255 |
-
"loss": 0.8659199714660645,
|
| 4256 |
-
"mean_token_accuracy": 0.8032226487994194,
|
| 4257 |
-
"num_tokens": 3173432.0,
|
| 4258 |
-
"step": 3830
|
| 4259 |
-
},
|
| 4260 |
-
{
|
| 4261 |
-
"entropy": 0.8979651927947998,
|
| 4262 |
-
"epoch": 0.6614132541015373,
|
| 4263 |
-
"grad_norm": 0.4457395672798157,
|
| 4264 |
-
"learning_rate": 6.775749224939719e-06,
|
| 4265 |
-
"loss": 0.8884675979614258,
|
| 4266 |
-
"mean_token_accuracy": 0.8004711583256722,
|
| 4267 |
-
"num_tokens": 3309359.0,
|
| 4268 |
-
"step": 3840
|
| 4269 |
-
},
|
| 4270 |
-
{
|
| 4271 |
-
"entropy": 0.8551001802086831,
|
| 4272 |
-
"epoch": 0.66313568445076,
|
| 4273 |
-
"grad_norm": 0.4345524311065674,
|
| 4274 |
-
"learning_rate": 6.741302101274544e-06,
|
| 4275 |
-
"loss": 0.8385509490966797,
|
| 4276 |
-
"mean_token_accuracy": 0.8087713450193406,
|
| 4277 |
-
"num_tokens": 3444443.0,
|
| 4278 |
-
"step": 3850
|
| 4279 |
-
},
|
| 4280 |
-
{
|
| 4281 |
-
"entropy": 0.891946230828762,
|
| 4282 |
-
"epoch": 0.6648581147999828,
|
| 4283 |
-
"grad_norm": 0.4427470862865448,
|
| 4284 |
-
"learning_rate": 6.7068549776093705e-06,
|
| 4285 |
-
"loss": 0.8853155136108398,
|
| 4286 |
-
"mean_token_accuracy": 0.7998019322752953,
|
| 4287 |
-
"num_tokens": 3582885.0,
|
| 4288 |
-
"step": 3860
|
| 4289 |
-
},
|
| 4290 |
-
{
|
| 4291 |
-
"entropy": 0.8642865002155304,
|
| 4292 |
-
"epoch": 0.6665805451492055,
|
| 4293 |
-
"grad_norm": 0.4064953327178955,
|
| 4294 |
-
"learning_rate": 6.672407853944196e-06,
|
| 4295 |
-
"loss": 0.8612741470336914,
|
| 4296 |
-
"mean_token_accuracy": 0.802153280377388,
|
| 4297 |
-
"num_tokens": 3722762.0,
|
| 4298 |
-
"step": 3870
|
| 4299 |
-
},
|
| 4300 |
-
{
|
| 4301 |
-
"entropy": 0.8654727056622505,
|
| 4302 |
-
"epoch": 0.6683029754984283,
|
| 4303 |
-
"grad_norm": 0.44759929180145264,
|
| 4304 |
-
"learning_rate": 6.637960730279023e-06,
|
| 4305 |
-
"loss": 0.8589719772338867,
|
| 4306 |
-
"mean_token_accuracy": 0.8055579513311386,
|
| 4307 |
-
"num_tokens": 3862098.0,
|
| 4308 |
-
"step": 3880
|
| 4309 |
-
},
|
| 4310 |
-
{
|
| 4311 |
-
"entropy": 0.9094881787896156,
|
| 4312 |
-
"epoch": 0.670025405847651,
|
| 4313 |
-
"grad_norm": 0.41999101638793945,
|
| 4314 |
-
"learning_rate": 6.603513606613848e-06,
|
| 4315 |
-
"loss": 0.901517391204834,
|
| 4316 |
-
"mean_token_accuracy": 0.798083683848381,
|
| 4317 |
-
"num_tokens": 4000549.0,
|
| 4318 |
-
"step": 3890
|
| 4319 |
-
},
|
| 4320 |
-
{
|
| 4321 |
-
"entropy": 0.9040753439068794,
|
| 4322 |
-
"epoch": 0.6717478361968738,
|
| 4323 |
-
"grad_norm": 0.39481016993522644,
|
| 4324 |
-
"learning_rate": 6.569066482948674e-06,
|
| 4325 |
-
"loss": 0.8963336944580078,
|
| 4326 |
-
"mean_token_accuracy": 0.7986732751131058,
|
| 4327 |
-
"num_tokens": 4137043.0,
|
| 4328 |
-
"step": 3900
|
| 4329 |
-
},
|
| 4330 |
-
{
|
| 4331 |
-
"epoch": 0.6717478361968738,
|
| 4332 |
-
"eval_entropy": 0.8978190375062579,
|
| 4333 |
-
"eval_loss": 0.8890023827552795,
|
| 4334 |
-
"eval_mean_token_accuracy": 0.8000849456530308,
|
| 4335 |
-
"eval_num_tokens": 4137043.0,
|
| 4336 |
-
"eval_runtime": 1997.0726,
|
| 4337 |
-
"eval_samples_per_second": 5.168,
|
| 4338 |
-
"eval_steps_per_second": 1.292,
|
| 4339 |
-
"step": 3900
|
| 4340 |
-
},
|
| 4341 |
-
{
|
| 4342 |
-
"entropy": 0.888623583316803,
|
| 4343 |
-
"epoch": 0.6734702665460965,
|
| 4344 |
-
"grad_norm": 0.3936791718006134,
|
| 4345 |
-
"learning_rate": 6.534619359283501e-06,
|
| 4346 |
-
"loss": 0.8851858139038086,
|
| 4347 |
-
"mean_token_accuracy": 0.80007184445858,
|
| 4348 |
-
"num_tokens": 4273008.0,
|
| 4349 |
-
"step": 3910
|
| 4350 |
-
},
|
| 4351 |
-
{
|
| 4352 |
-
"entropy": 0.9054401487112045,
|
| 4353 |
-
"epoch": 0.6751926968953192,
|
| 4354 |
-
"grad_norm": 0.4450265169143677,
|
| 4355 |
-
"learning_rate": 6.500172235618326e-06,
|
| 4356 |
-
"loss": 0.8935989379882813,
|
| 4357 |
-
"mean_token_accuracy": 0.7988793134689331,
|
| 4358 |
-
"num_tokens": 4415036.0,
|
| 4359 |
-
"step": 3920
|
| 4360 |
-
},
|
| 4361 |
-
{
|
| 4362 |
-
"entropy": 0.8839291751384735,
|
| 4363 |
-
"epoch": 0.6769151272445421,
|
| 4364 |
-
"grad_norm": 0.44173672795295715,
|
| 4365 |
-
"learning_rate": 6.4657251119531525e-06,
|
| 4366 |
-
"loss": 0.8775888442993164,
|
| 4367 |
-
"mean_token_accuracy": 0.8007264167070389,
|
| 4368 |
-
"num_tokens": 4552708.0,
|
| 4369 |
-
"step": 3930
|
| 4370 |
-
},
|
| 4371 |
-
{
|
| 4372 |
-
"entropy": 0.8589925251901149,
|
| 4373 |
-
"epoch": 0.6786375575937648,
|
| 4374 |
-
"grad_norm": 0.3620625436306,
|
| 4375 |
-
"learning_rate": 6.4312779882879784e-06,
|
| 4376 |
-
"loss": 0.8551578521728516,
|
| 4377 |
-
"mean_token_accuracy": 0.8064107894897461,
|
| 4378 |
-
"num_tokens": 4696290.0,
|
| 4379 |
-
"step": 3940
|
| 4380 |
-
},
|
| 4381 |
-
{
|
| 4382 |
-
"entropy": 0.8445931300520897,
|
| 4383 |
-
"epoch": 0.6803599879429876,
|
| 4384 |
-
"grad_norm": 0.4519362449645996,
|
| 4385 |
-
"learning_rate": 6.396830864622805e-06,
|
| 4386 |
-
"loss": 0.8345125198364258,
|
| 4387 |
-
"mean_token_accuracy": 0.8092260167002678,
|
| 4388 |
-
"num_tokens": 4832087.0,
|
| 4389 |
-
"step": 3950
|
| 4390 |
-
},
|
| 4391 |
-
{
|
| 4392 |
-
"entropy": 0.9084868378937244,
|
| 4393 |
-
"epoch": 0.6820824182922103,
|
| 4394 |
-
"grad_norm": 0.40370234847068787,
|
| 4395 |
-
"learning_rate": 6.36238374095763e-06,
|
| 4396 |
-
"loss": 0.901630973815918,
|
| 4397 |
-
"mean_token_accuracy": 0.7952028334140777,
|
| 4398 |
-
"num_tokens": 4972261.0,
|
| 4399 |
-
"step": 3960
|
| 4400 |
-
},
|
| 4401 |
-
{
|
| 4402 |
-
"entropy": 0.8734813243150711,
|
| 4403 |
-
"epoch": 0.6838048486414331,
|
| 4404 |
-
"grad_norm": 0.4129689335823059,
|
| 4405 |
-
"learning_rate": 6.327936617292456e-06,
|
| 4406 |
-
"loss": 0.8525594711303711,
|
| 4407 |
-
"mean_token_accuracy": 0.806535467505455,
|
| 4408 |
-
"num_tokens": 5105943.0,
|
| 4409 |
-
"step": 3970
|
| 4410 |
-
},
|
| 4411 |
-
{
|
| 4412 |
-
"entropy": 0.8367196694016457,
|
| 4413 |
-
"epoch": 0.6855272789906558,
|
| 4414 |
-
"grad_norm": 0.38863131403923035,
|
| 4415 |
-
"learning_rate": 6.293489493627283e-06,
|
| 4416 |
-
"loss": 0.8323159217834473,
|
| 4417 |
-
"mean_token_accuracy": 0.8115260645747184,
|
| 4418 |
-
"num_tokens": 5241657.0,
|
| 4419 |
-
"step": 3980
|
| 4420 |
-
},
|
| 4421 |
-
{
|
| 4422 |
-
"entropy": 0.8844600863754749,
|
| 4423 |
-
"epoch": 0.6872497093398786,
|
| 4424 |
-
"grad_norm": 0.4728356599807739,
|
| 4425 |
-
"learning_rate": 6.259042369962109e-06,
|
| 4426 |
-
"loss": 0.8816679000854493,
|
| 4427 |
-
"mean_token_accuracy": 0.800590705871582,
|
| 4428 |
-
"num_tokens": 5381973.0,
|
| 4429 |
-
"step": 3990
|
| 4430 |
-
},
|
| 4431 |
-
{
|
| 4432 |
-
"entropy": 0.865593820810318,
|
| 4433 |
-
"epoch": 0.6889721396891013,
|
| 4434 |
-
"grad_norm": 0.3811900317668915,
|
| 4435 |
-
"learning_rate": 6.224595246296935e-06,
|
| 4436 |
-
"loss": 0.8632452011108398,
|
| 4437 |
-
"mean_token_accuracy": 0.8051433145999909,
|
| 4438 |
-
"num_tokens": 5520841.0,
|
| 4439 |
-
"step": 4000
|
| 4440 |
-
},
|
| 4441 |
-
{
|
| 4442 |
-
"epoch": 0.6889721396891013,
|
| 4443 |
-
"eval_entropy": 0.8943157961367637,
|
| 4444 |
-
"eval_loss": 0.8887593150138855,
|
| 4445 |
-
"eval_mean_token_accuracy": 0.8001742187002656,
|
| 4446 |
-
"eval_num_tokens": 5520841.0,
|
| 4447 |
-
"eval_runtime": 1998.5827,
|
| 4448 |
-
"eval_samples_per_second": 5.164,
|
| 4449 |
-
"eval_steps_per_second": 1.291,
|
| 4450 |
-
"step": 4000
|
| 4451 |
}
|
| 4452 |
],
|
| 4453 |
"logging_steps": 10,
|
|
@@ -4467,7 +4023,7 @@
|
|
| 4467 |
"attributes": {}
|
| 4468 |
}
|
| 4469 |
},
|
| 4470 |
-
"total_flos":
|
| 4471 |
"train_batch_size": 2,
|
| 4472 |
"trial_name": null,
|
| 4473 |
"trial_params": null
|
|
|
|
| 2 |
"best_global_step": null,
|
| 3 |
"best_metric": null,
|
| 4 |
"best_model_checkpoint": null,
|
| 5 |
+
"epoch": 0.6200749257201912,
|
| 6 |
"eval_steps": 100,
|
| 7 |
+
"global_step": 3600,
|
| 8 |
"is_hyper_param_search": false,
|
| 9 |
"is_local_process_zero": true,
|
| 10 |
"is_world_process_zero": true,
|
|
|
|
| 4004 |
"eval_samples_per_second": 5.088,
|
| 4005 |
"eval_steps_per_second": 1.272,
|
| 4006 |
"step": 3600
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4007 |
}
|
| 4008 |
],
|
| 4009 |
"logging_steps": 10,
|
|
|
|
| 4023 |
"attributes": {}
|
| 4024 |
}
|
| 4025 |
},
|
| 4026 |
+
"total_flos": 5.912574753998111e+17,
|
| 4027 |
"train_batch_size": 2,
|
| 4028 |
"trial_name": null,
|
| 4029 |
"trial_params": null
|