| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.019913376810872702, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.00019913376810872703, | |
| "eval_loss": 2.2963063716888428, | |
| "eval_runtime": 622.897, | |
| "eval_samples_per_second": 13.578, | |
| "eval_steps_per_second": 1.699, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0005974013043261811, | |
| "grad_norm": 23.913619995117188, | |
| "learning_rate": 1.5e-05, | |
| "loss": 8.8941, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0011948026086523623, | |
| "grad_norm": 19.30548858642578, | |
| "learning_rate": 3e-05, | |
| "loss": 8.8968, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0017922039129785433, | |
| "grad_norm": 10.14571762084961, | |
| "learning_rate": 4.5e-05, | |
| "loss": 8.4803, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0017922039129785433, | |
| "eval_loss": 1.8854235410690308, | |
| "eval_runtime": 627.0021, | |
| "eval_samples_per_second": 13.49, | |
| "eval_steps_per_second": 1.687, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0023896052173047245, | |
| "grad_norm": 5.248955249786377, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 7.1413, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0029870065216309055, | |
| "grad_norm": 5.370950222015381, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 6.8965, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0035844078259570865, | |
| "grad_norm": 6.065382480621338, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 7.1409, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0035844078259570865, | |
| "eval_loss": 1.7099746465682983, | |
| "eval_runtime": 626.7094, | |
| "eval_samples_per_second": 13.496, | |
| "eval_steps_per_second": 1.688, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0041818091302832676, | |
| "grad_norm": 4.529496669769287, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 7.241, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.004779210434609449, | |
| "grad_norm": 3.878418207168579, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 6.9688, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.00537661173893563, | |
| "grad_norm": 5.248115539550781, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 6.7016, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.00537661173893563, | |
| "eval_loss": 1.6524488925933838, | |
| "eval_runtime": 627.2362, | |
| "eval_samples_per_second": 13.485, | |
| "eval_steps_per_second": 1.687, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.005974013043261811, | |
| "grad_norm": 5.091797828674316, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 6.703, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0065714143475879925, | |
| "grad_norm": 4.741832256317139, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 5.8822, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.007168815651914173, | |
| "grad_norm": 4.389284133911133, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 6.637, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.007168815651914173, | |
| "eval_loss": 1.626313328742981, | |
| "eval_runtime": 627.4338, | |
| "eval_samples_per_second": 13.48, | |
| "eval_steps_per_second": 1.686, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0077662169562403545, | |
| "grad_norm": 3.989469528198242, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 6.5048, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.008363618260566535, | |
| "grad_norm": 4.279814720153809, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 6.5277, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.008961019564892717, | |
| "grad_norm": 3.723292827606201, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 6.2794, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.008961019564892717, | |
| "eval_loss": 1.6118539571762085, | |
| "eval_runtime": 626.9775, | |
| "eval_samples_per_second": 13.49, | |
| "eval_steps_per_second": 1.687, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.009558420869218898, | |
| "grad_norm": 3.986083507537842, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 6.0171, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.01015582217354508, | |
| "grad_norm": 4.378915309906006, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 6.5238, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.01075322347787126, | |
| "grad_norm": 4.291739463806152, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 6.2643, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.01075322347787126, | |
| "eval_loss": 1.6048556566238403, | |
| "eval_runtime": 627.3604, | |
| "eval_samples_per_second": 13.482, | |
| "eval_steps_per_second": 1.686, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.01135062478219744, | |
| "grad_norm": 4.052128314971924, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 6.4014, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.011948026086523622, | |
| "grad_norm": 4.348297119140625, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 6.6049, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.012545427390849804, | |
| "grad_norm": 4.070796012878418, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 6.6138, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.012545427390849804, | |
| "eval_loss": 1.5996973514556885, | |
| "eval_runtime": 627.0719, | |
| "eval_samples_per_second": 13.488, | |
| "eval_steps_per_second": 1.687, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.013142828695175985, | |
| "grad_norm": 3.7099554538726807, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 6.5339, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.013740229999502166, | |
| "grad_norm": 4.7017998695373535, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 6.5618, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.014337631303828346, | |
| "grad_norm": 3.729522705078125, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 6.0646, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.014337631303828346, | |
| "eval_loss": 1.5968443155288696, | |
| "eval_runtime": 627.1273, | |
| "eval_samples_per_second": 13.487, | |
| "eval_steps_per_second": 1.687, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.014935032608154528, | |
| "grad_norm": 3.717174768447876, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 6.2037, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.015532433912480709, | |
| "grad_norm": 5.319967746734619, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 6.2515, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.01612983521680689, | |
| "grad_norm": 4.202660083770752, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 6.0643, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.01612983521680689, | |
| "eval_loss": 1.5953447818756104, | |
| "eval_runtime": 627.5483, | |
| "eval_samples_per_second": 13.478, | |
| "eval_steps_per_second": 1.686, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.01672723652113307, | |
| "grad_norm": 4.394832134246826, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 6.6288, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.017324637825459253, | |
| "grad_norm": 5.086785316467285, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 6.3303, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.017922039129785433, | |
| "grad_norm": 3.8469746112823486, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 6.6755, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.017922039129785433, | |
| "eval_loss": 1.5948328971862793, | |
| "eval_runtime": 627.4465, | |
| "eval_samples_per_second": 13.48, | |
| "eval_steps_per_second": 1.686, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.018519440434111613, | |
| "grad_norm": 4.4008331298828125, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 6.3269, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.019116841738437796, | |
| "grad_norm": 3.7891712188720703, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 6.312, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.019714243042763976, | |
| "grad_norm": 4.460823059082031, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 6.0766, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.019714243042763976, | |
| "eval_loss": 1.5946401357650757, | |
| "eval_runtime": 626.8725, | |
| "eval_samples_per_second": 13.492, | |
| "eval_steps_per_second": 1.688, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.484255733649244e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |