| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 3.0, |
| "eval_steps": 50, |
| "global_step": 225, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.06711409395973154, |
| "grad_norm": 771.0387290625706, |
| "learning_rate": 6.666666666666667e-06, |
| "loss": 3.3403587341308594, |
| "step": 5 |
| }, |
| { |
| "epoch": 0.1342281879194631, |
| "grad_norm": 24.798543146738286, |
| "learning_rate": 1.5000000000000002e-05, |
| "loss": 1.2873397827148438, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.20134228187919462, |
| "grad_norm": 9.663013484017599, |
| "learning_rate": 1.9995649501315172e-05, |
| "loss": 0.29829452037811277, |
| "step": 15 |
| }, |
| { |
| "epoch": 0.2684563758389262, |
| "grad_norm": 5.6477057523539, |
| "learning_rate": 1.9946749853292233e-05, |
| "loss": 0.1349006175994873, |
| "step": 20 |
| }, |
| { |
| "epoch": 0.33557046979865773, |
| "grad_norm": 4.057098021702121, |
| "learning_rate": 1.9843779142227258e-05, |
| "loss": 0.07990359663963317, |
| "step": 25 |
| }, |
| { |
| "epoch": 0.40268456375838924, |
| "grad_norm": 1.914170020073957, |
| "learning_rate": 1.9687297122400952e-05, |
| "loss": 0.03623477220535278, |
| "step": 30 |
| }, |
| { |
| "epoch": 0.4697986577181208, |
| "grad_norm": 0.05978450488009304, |
| "learning_rate": 1.947815443842424e-05, |
| "loss": 0.0021600183099508286, |
| "step": 35 |
| }, |
| { |
| "epoch": 0.5369127516778524, |
| "grad_norm": 0.012301196399380163, |
| "learning_rate": 1.9217488001088784e-05, |
| "loss": 0.00010623512789607049, |
| "step": 40 |
| }, |
| { |
| "epoch": 0.6040268456375839, |
| "grad_norm": 0.004193028483875623, |
| "learning_rate": 1.8906714807059218e-05, |
| "loss": 2.885170979425311e-05, |
| "step": 45 |
| }, |
| { |
| "epoch": 0.6711409395973155, |
| "grad_norm": 0.0021696829162455526, |
| "learning_rate": 1.8547524236003675e-05, |
| "loss": 1.2394384248182177e-05, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.6711409395973155, |
| "eval_loss": 9.160361514659598e-06, |
| "eval_runtime": 2.223, |
| "eval_samples_per_second": 112.459, |
| "eval_steps_per_second": 7.197, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.738255033557047, |
| "grad_norm": 0.001552062897034346, |
| "learning_rate": 1.8141868867035745e-05, |
| "loss": 7.670311606489123e-06, |
| "step": 55 |
| }, |
| { |
| "epoch": 0.8053691275167785, |
| "grad_norm": 0.0012379335033009154, |
| "learning_rate": 1.7691953864390208e-05, |
| "loss": 5.718640022678301e-06, |
| "step": 60 |
| }, |
| { |
| "epoch": 0.87248322147651, |
| "grad_norm": 0.0010541000193795753, |
| "learning_rate": 1.7200224990032577e-05, |
| "loss": 4.781161987921223e-06, |
| "step": 65 |
| }, |
| { |
| "epoch": 0.9395973154362416, |
| "grad_norm": 0.000928873490914045, |
| "learning_rate": 1.666935530836651e-05, |
| "loss": 4.1388742829440165e-06, |
| "step": 70 |
| }, |
| { |
| "epoch": 1.0, |
| "grad_norm": 0.0008922235905923142, |
| "learning_rate": 1.6102230655313076e-05, |
| "loss": 3.85485400329344e-06, |
| "step": 75 |
| }, |
| { |
| "epoch": 1.0671140939597314, |
| "grad_norm": 0.0008075855123012748, |
| "learning_rate": 1.5501933950752655e-05, |
| "loss": 3.514360651024617e-06, |
| "step": 80 |
| }, |
| { |
| "epoch": 1.1342281879194631, |
| "grad_norm": 0.0007917095955772439, |
| "learning_rate": 1.4871728439607967e-05, |
| "loss": 3.3707532566040753e-06, |
| "step": 85 |
| }, |
| { |
| "epoch": 1.2013422818791946, |
| "grad_norm": 0.0007942119288409565, |
| "learning_rate": 1.4215039952670482e-05, |
| "loss": 3.3559525036253033e-06, |
| "step": 90 |
| }, |
| { |
| "epoch": 1.2684563758389262, |
| "grad_norm": 0.0007667582632583201, |
| "learning_rate": 1.3535438283601437e-05, |
| "loss": 3.262608515797183e-06, |
| "step": 95 |
| }, |
| { |
| "epoch": 1.3355704697986577, |
| "grad_norm": 0.000737964470158104, |
| "learning_rate": 1.2836617783342968e-05, |
| "loss": 3.128412936348468e-06, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.3355704697986577, |
| "eval_loss": 3.1148604193731444e-06, |
| "eval_runtime": 2.0231, |
| "eval_samples_per_second": 123.573, |
| "eval_steps_per_second": 7.909, |
| "step": 100 |
| }, |
| { |
| "epoch": 1.4026845637583891, |
| "grad_norm": 0.0007199615576756328, |
| "learning_rate": 1.2122377277429231e-05, |
| "loss": 3.0415885703405365e-06, |
| "step": 105 |
| }, |
| { |
| "epoch": 1.4697986577181208, |
| "grad_norm": 0.0007126247175517023, |
| "learning_rate": 1.1396599415368062e-05, |
| "loss": 2.98672603094019e-06, |
| "step": 110 |
| }, |
| { |
| "epoch": 1.5369127516778525, |
| "grad_norm": 0.0007147228786157012, |
| "learning_rate": 1.066322956435104e-05, |
| "loss": 2.9487675419659355e-06, |
| "step": 115 |
| }, |
| { |
| "epoch": 1.604026845637584, |
| "grad_norm": 0.000710730245597985, |
| "learning_rate": 9.926254362026875e-06, |
| "loss": 2.942025457741693e-06, |
| "step": 120 |
| }, |
| { |
| "epoch": 1.6711409395973154, |
| "grad_norm": 0.0007130604665711499, |
| "learning_rate": 9.189680044926402e-06, |
| "loss": 2.926953311543912e-06, |
| "step": 125 |
| }, |
| { |
| "epoch": 1.738255033557047, |
| "grad_norm": 0.0007082339750801266, |
| "learning_rate": 8.457510670346976e-06, |
| "loss": 2.9204267775639893e-06, |
| "step": 130 |
| }, |
| { |
| "epoch": 1.8053691275167785, |
| "grad_norm": 0.0007212567260949268, |
| "learning_rate": 7.733726350083331e-06, |
| "loss": 2.9483762773452327e-06, |
| "step": 135 |
| }, |
| { |
| "epoch": 1.87248322147651, |
| "grad_norm": 0.0007020476633877951, |
| "learning_rate": 7.022261614327448e-06, |
| "loss": 5.818682257086038e-06, |
| "step": 140 |
| }, |
| { |
| "epoch": 1.9395973154362416, |
| "grad_norm": 0.0006973774426251346, |
| "learning_rate": 6.326984023352435e-06, |
| "loss": 2.8704187570838256e-06, |
| "step": 145 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 0.000682624605940799, |
| "learning_rate": 5.651673143248509e-06, |
| "loss": 2.8186999770696274e-06, |
| "step": 150 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_loss": 2.8178355933050625e-06, |
| "eval_runtime": 2.127, |
| "eval_samples_per_second": 117.535, |
| "eval_steps_per_second": 7.522, |
| "step": 150 |
| }, |
| { |
| "epoch": 2.0671140939597317, |
| "grad_norm": 0.0006879802420483138, |
| "learning_rate": 5.000000000000003e-06, |
| "loss": 2.7831285478896463e-06, |
| "step": 155 |
| }, |
| { |
| "epoch": 2.134228187919463, |
| "grad_norm": 0.0006632420952139857, |
| "learning_rate": 4.375507123592194e-06, |
| "loss": 2.7615027647698297e-06, |
| "step": 160 |
| }, |
| { |
| "epoch": 2.2013422818791946, |
| "grad_norm": 0.0006656627025681546, |
| "learning_rate": 3.781589290629266e-06, |
| "loss": 2.7448764740256593e-06, |
| "step": 165 |
| }, |
| { |
| "epoch": 2.2684563758389262, |
| "grad_norm": 0.0006705715879794475, |
| "learning_rate": 3.2214750701474875e-06, |
| "loss": 2.7252637664787473e-06, |
| "step": 170 |
| }, |
| { |
| "epoch": 2.335570469798658, |
| "grad_norm": 0.000668328323821757, |
| "learning_rate": 2.698209272941659e-06, |
| "loss": 2.7195979782845826e-06, |
| "step": 175 |
| }, |
| { |
| "epoch": 2.402684563758389, |
| "grad_norm": 0.0006694753608437758, |
| "learning_rate": 2.2146363998111077e-06, |
| "loss": 2.7082156520918944e-06, |
| "step": 180 |
| }, |
| { |
| "epoch": 2.469798657718121, |
| "grad_norm": 0.0006609888028080037, |
| "learning_rate": 1.773385178701561e-06, |
| "loss": 2.6918616640614347e-06, |
| "step": 185 |
| }, |
| { |
| "epoch": 2.5369127516778525, |
| "grad_norm": 0.0006616546862766573, |
| "learning_rate": 1.3768542747997215e-06, |
| "loss": 2.6803711079992356e-06, |
| "step": 190 |
| }, |
| { |
| "epoch": 2.604026845637584, |
| "grad_norm": 0.0006557050922852907, |
| "learning_rate": 1.0271992512613239e-06, |
| "loss": 2.687972482817713e-06, |
| "step": 195 |
| }, |
| { |
| "epoch": 2.6711409395973154, |
| "grad_norm": 0.0006564030982639905, |
| "learning_rate": 7.263208514547548e-07, |
| "loss": 2.6694764528656378e-06, |
| "step": 200 |
| }, |
| { |
| "epoch": 2.6711409395973154, |
| "eval_loss": 2.682836338863126e-06, |
| "eval_runtime": 2.0484, |
| "eval_samples_per_second": 122.044, |
| "eval_steps_per_second": 7.811, |
| "step": 200 |
| }, |
| { |
| "epoch": 2.738255033557047, |
| "grad_norm": 0.0006632458099436633, |
| "learning_rate": 4.7585466641868696e-07, |
| "loss": 2.6877429263549858e-06, |
| "step": 205 |
| }, |
| { |
| "epoch": 2.8053691275167782, |
| "grad_norm": 0.0006591045490234142, |
| "learning_rate": 2.7716224370188836e-07, |
| "loss": 2.6864749088417738e-06, |
| "step": 210 |
| }, |
| { |
| "epoch": 2.87248322147651, |
| "grad_norm": 0.0006574142715557724, |
| "learning_rate": 1.3132368591809553e-07, |
| "loss": 2.678953023860231e-06, |
| "step": 215 |
| }, |
| { |
| "epoch": 2.9395973154362416, |
| "grad_norm": 0.0006592814820222594, |
| "learning_rate": 3.913177925055189e-08, |
| "loss": 2.6803716536960563e-06, |
| "step": 220 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 0.0006633342417694629, |
| "learning_rate": 1.0876838240125865e-09, |
| "loss": 2.679282988538034e-06, |
| "step": 225 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_loss": 2.67330960923573e-06, |
| "eval_runtime": 2.091, |
| "eval_samples_per_second": 119.56, |
| "eval_steps_per_second": 7.652, |
| "step": 225 |
| }, |
| { |
| "epoch": 3.0, |
| "step": 225, |
| "total_flos": 4338410323968.0, |
| "train_loss": 0.1150990028197177, |
| "train_runtime": 1408.4559, |
| "train_samples_per_second": 10.117, |
| "train_steps_per_second": 0.16 |
| } |
| ], |
| "logging_steps": 5, |
| "max_steps": 225, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 3, |
| "save_steps": 50, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 4338410323968.0, |
| "train_batch_size": 4, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
