dada22231's picture
Training in progress, step 50, checkpoint
b18ded5 verified
raw
history blame
9.96 kB
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.5392652510953826,
"eval_steps": 25,
"global_step": 50,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.010785305021907651,
"grad_norm": 4.344828128814697,
"learning_rate": 5e-05,
"loss": 7.3561,
"step": 1
},
{
"epoch": 0.010785305021907651,
"eval_loss": 4.603348731994629,
"eval_runtime": 2.6307,
"eval_samples_per_second": 19.006,
"eval_steps_per_second": 4.942,
"step": 1
},
{
"epoch": 0.021570610043815303,
"grad_norm": 5.045979022979736,
"learning_rate": 0.0001,
"loss": 6.7432,
"step": 2
},
{
"epoch": 0.032355915065722954,
"grad_norm": 4.547981262207031,
"learning_rate": 9.990365154573717e-05,
"loss": 6.1515,
"step": 3
},
{
"epoch": 0.043141220087630605,
"grad_norm": 3.4077675342559814,
"learning_rate": 9.961501876182148e-05,
"loss": 5.4805,
"step": 4
},
{
"epoch": 0.053926525109538256,
"grad_norm": 2.6199681758880615,
"learning_rate": 9.913533761814537e-05,
"loss": 4.875,
"step": 5
},
{
"epoch": 0.06471183013144591,
"grad_norm": 2.0004589557647705,
"learning_rate": 9.846666218300807e-05,
"loss": 4.2886,
"step": 6
},
{
"epoch": 0.07549713515335356,
"grad_norm": 1.5946983098983765,
"learning_rate": 9.761185582727977e-05,
"loss": 3.8103,
"step": 7
},
{
"epoch": 0.08628244017526121,
"grad_norm": 1.276253342628479,
"learning_rate": 9.657457896300791e-05,
"loss": 3.3543,
"step": 8
},
{
"epoch": 0.09706774519716886,
"grad_norm": 1.0471776723861694,
"learning_rate": 9.535927336897098e-05,
"loss": 2.954,
"step": 9
},
{
"epoch": 0.10785305021907651,
"grad_norm": 0.9004111289978027,
"learning_rate": 9.397114317029975e-05,
"loss": 2.7611,
"step": 10
},
{
"epoch": 0.11863835524098416,
"grad_norm": 0.5792182087898254,
"learning_rate": 9.241613255361455e-05,
"loss": 2.666,
"step": 11
},
{
"epoch": 0.12942366026289182,
"grad_norm": 0.24996623396873474,
"learning_rate": 9.070090031310558e-05,
"loss": 2.3539,
"step": 12
},
{
"epoch": 0.14020896528479945,
"grad_norm": 3.6459882259368896,
"learning_rate": 8.883279133655399e-05,
"loss": 5.8301,
"step": 13
},
{
"epoch": 0.15099427030670712,
"grad_norm": 2.7042088508605957,
"learning_rate": 8.681980515339464e-05,
"loss": 5.3308,
"step": 14
},
{
"epoch": 0.16177957532861476,
"grad_norm": 2.0865371227264404,
"learning_rate": 8.467056167950311e-05,
"loss": 4.742,
"step": 15
},
{
"epoch": 0.17256488035052242,
"grad_norm": 1.6845545768737793,
"learning_rate": 8.239426430539243e-05,
"loss": 4.199,
"step": 16
},
{
"epoch": 0.18335018537243006,
"grad_norm": 1.4420006275177002,
"learning_rate": 8.000066048588211e-05,
"loss": 3.8033,
"step": 17
},
{
"epoch": 0.19413549039433772,
"grad_norm": 1.2050021886825562,
"learning_rate": 7.75e-05,
"loss": 3.4628,
"step": 18
},
{
"epoch": 0.20492079541624536,
"grad_norm": 1.108812689781189,
"learning_rate": 7.490299105985507e-05,
"loss": 3.1717,
"step": 19
},
{
"epoch": 0.21570610043815303,
"grad_norm": 0.9539301991462708,
"learning_rate": 7.222075445642904e-05,
"loss": 2.8746,
"step": 20
},
{
"epoch": 0.22649140546006066,
"grad_norm": 0.5657325387001038,
"learning_rate": 6.946477593864228e-05,
"loss": 2.7393,
"step": 21
},
{
"epoch": 0.23727671048196833,
"grad_norm": 0.4725482761859894,
"learning_rate": 6.664685702961344e-05,
"loss": 2.5845,
"step": 22
},
{
"epoch": 0.24806201550387597,
"grad_norm": 0.5372779965400696,
"learning_rate": 6.377906449072578e-05,
"loss": 2.6176,
"step": 23
},
{
"epoch": 0.25884732052578363,
"grad_norm": 0.42209765315055847,
"learning_rate": 6.087367864990233e-05,
"loss": 2.391,
"step": 24
},
{
"epoch": 0.2696326255476913,
"grad_norm": 0.32752636075019836,
"learning_rate": 5.794314081535644e-05,
"loss": 2.4172,
"step": 25
},
{
"epoch": 0.2696326255476913,
"eval_loss": 3.2214887142181396,
"eval_runtime": 2.2412,
"eval_samples_per_second": 22.31,
"eval_steps_per_second": 5.801,
"step": 25
},
{
"epoch": 0.2804179305695989,
"grad_norm": 2.680741548538208,
"learning_rate": 5.500000000000001e-05,
"loss": 4.7656,
"step": 26
},
{
"epoch": 0.29120323559150657,
"grad_norm": 1.608992099761963,
"learning_rate": 5.205685918464356e-05,
"loss": 4.0335,
"step": 27
},
{
"epoch": 0.30198854061341424,
"grad_norm": 1.2295485734939575,
"learning_rate": 4.912632135009769e-05,
"loss": 3.6691,
"step": 28
},
{
"epoch": 0.3127738456353219,
"grad_norm": 0.9207251071929932,
"learning_rate": 4.6220935509274235e-05,
"loss": 3.3744,
"step": 29
},
{
"epoch": 0.3235591506572295,
"grad_norm": 1.0553677082061768,
"learning_rate": 4.3353142970386564e-05,
"loss": 3.0976,
"step": 30
},
{
"epoch": 0.3343444556791372,
"grad_norm": 1.3266836404800415,
"learning_rate": 4.053522406135775e-05,
"loss": 2.8384,
"step": 31
},
{
"epoch": 0.34512976070104484,
"grad_norm": 1.3463107347488403,
"learning_rate": 3.777924554357096e-05,
"loss": 2.6958,
"step": 32
},
{
"epoch": 0.35591506572295245,
"grad_norm": 1.297961711883545,
"learning_rate": 3.509700894014496e-05,
"loss": 2.6287,
"step": 33
},
{
"epoch": 0.3667003707448601,
"grad_norm": 1.0394666194915771,
"learning_rate": 3.250000000000001e-05,
"loss": 2.5336,
"step": 34
},
{
"epoch": 0.3774856757667678,
"grad_norm": 0.8954881429672241,
"learning_rate": 2.9999339514117912e-05,
"loss": 2.5946,
"step": 35
},
{
"epoch": 0.38827098078867545,
"grad_norm": 0.7060102224349976,
"learning_rate": 2.760573569460757e-05,
"loss": 2.4813,
"step": 36
},
{
"epoch": 0.39905628581058306,
"grad_norm": 0.40887507796287537,
"learning_rate": 2.53294383204969e-05,
"loss": 2.3328,
"step": 37
},
{
"epoch": 0.4098415908324907,
"grad_norm": 2.2469213008880615,
"learning_rate": 2.3180194846605367e-05,
"loss": 4.5164,
"step": 38
},
{
"epoch": 0.4206268958543984,
"grad_norm": 2.1526901721954346,
"learning_rate": 2.1167208663446025e-05,
"loss": 3.9944,
"step": 39
},
{
"epoch": 0.43141220087630605,
"grad_norm": 1.8147079944610596,
"learning_rate": 1.9299099686894423e-05,
"loss": 3.5679,
"step": 40
},
{
"epoch": 0.44219750589821366,
"grad_norm": 1.4103679656982422,
"learning_rate": 1.758386744638546e-05,
"loss": 3.2538,
"step": 41
},
{
"epoch": 0.4529828109201213,
"grad_norm": 1.1426249742507935,
"learning_rate": 1.602885682970026e-05,
"loss": 2.9928,
"step": 42
},
{
"epoch": 0.463768115942029,
"grad_norm": 0.9264675974845886,
"learning_rate": 1.464072663102903e-05,
"loss": 2.8695,
"step": 43
},
{
"epoch": 0.47455342096393666,
"grad_norm": 0.6036688089370728,
"learning_rate": 1.3425421036992098e-05,
"loss": 2.7405,
"step": 44
},
{
"epoch": 0.48533872598584427,
"grad_norm": 0.3461301028728485,
"learning_rate": 1.2388144172720251e-05,
"loss": 2.5788,
"step": 45
},
{
"epoch": 0.49612403100775193,
"grad_norm": 0.3648104667663574,
"learning_rate": 1.1533337816991932e-05,
"loss": 2.5385,
"step": 46
},
{
"epoch": 0.5069093360296596,
"grad_norm": 0.40517136454582214,
"learning_rate": 1.0864662381854632e-05,
"loss": 2.495,
"step": 47
},
{
"epoch": 0.5176946410515673,
"grad_norm": 0.41946572065353394,
"learning_rate": 1.0384981238178534e-05,
"loss": 2.5453,
"step": 48
},
{
"epoch": 0.5284799460734749,
"grad_norm": 0.4738365709781647,
"learning_rate": 1.0096348454262845e-05,
"loss": 2.3179,
"step": 49
},
{
"epoch": 0.5392652510953826,
"grad_norm": 0.3218916356563568,
"learning_rate": 1e-05,
"loss": 2.4055,
"step": 50
},
{
"epoch": 0.5392652510953826,
"eval_loss": 2.980289936065674,
"eval_runtime": 2.2402,
"eval_samples_per_second": 22.32,
"eval_steps_per_second": 5.803,
"step": 50
}
],
"logging_steps": 1,
"max_steps": 50,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 25,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 1.5594001969184768e+17,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}