| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.1968503937007874, | |
| "eval_steps": 500, | |
| "global_step": 300, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 9.97353458404541, | |
| "learning_rate": 2.424749163879599e-05, | |
| "loss": 2.1561, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 10.946741104125977, | |
| "learning_rate": 2.3411371237458197e-05, | |
| "loss": 1.6451, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 10.761913299560547, | |
| "learning_rate": 2.2575250836120402e-05, | |
| "loss": 1.4919, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 15.36636734008789, | |
| "learning_rate": 2.173913043478261e-05, | |
| "loss": 1.5464, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 9.350459098815918, | |
| "learning_rate": 2.090301003344482e-05, | |
| "loss": 1.5388, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 10.499181747436523, | |
| "learning_rate": 2.0066889632107023e-05, | |
| "loss": 1.5963, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 10.65820598602295, | |
| "learning_rate": 1.923076923076923e-05, | |
| "loss": 1.5128, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 9.002087593078613, | |
| "learning_rate": 1.8394648829431436e-05, | |
| "loss": 1.609, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 8.280915260314941, | |
| "learning_rate": 1.7558528428093644e-05, | |
| "loss": 1.4611, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 8.664236068725586, | |
| "learning_rate": 1.6722408026755853e-05, | |
| "loss": 1.4765, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 8.205462455749512, | |
| "learning_rate": 1.588628762541806e-05, | |
| "loss": 1.3903, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 8.982340812683105, | |
| "learning_rate": 1.5050167224080269e-05, | |
| "loss": 1.3134, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 7.957040786743164, | |
| "learning_rate": 1.4214046822742474e-05, | |
| "loss": 1.3256, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 8.92203426361084, | |
| "learning_rate": 1.3377926421404682e-05, | |
| "loss": 1.5535, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 6.650272369384766, | |
| "learning_rate": 1.254180602006689e-05, | |
| "loss": 1.5439, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 12.875391006469727, | |
| "learning_rate": 1.1705685618729099e-05, | |
| "loss": 1.3592, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 11.99698257446289, | |
| "learning_rate": 1.0869565217391305e-05, | |
| "loss": 1.4613, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 6.687454700469971, | |
| "learning_rate": 1.0033444816053512e-05, | |
| "loss": 1.2539, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 7.108691692352295, | |
| "learning_rate": 9.197324414715718e-06, | |
| "loss": 1.4818, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 7.420890808105469, | |
| "learning_rate": 8.361204013377926e-06, | |
| "loss": 1.332, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 7.554871082305908, | |
| "learning_rate": 7.5250836120401346e-06, | |
| "loss": 1.3705, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 8.120746612548828, | |
| "learning_rate": 6.688963210702341e-06, | |
| "loss": 1.3652, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 7.490564823150635, | |
| "learning_rate": 5.852842809364549e-06, | |
| "loss": 1.3594, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 9.032505989074707, | |
| "learning_rate": 5.016722408026756e-06, | |
| "loss": 1.2791, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 7.20292329788208, | |
| "learning_rate": 4.180602006688963e-06, | |
| "loss": 1.4569, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 9.031627655029297, | |
| "learning_rate": 3.3444816053511705e-06, | |
| "loss": 1.3965, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 7.822471618652344, | |
| "learning_rate": 2.508361204013378e-06, | |
| "loss": 1.418, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 15.259939193725586, | |
| "learning_rate": 1.6722408026755853e-06, | |
| "loss": 1.3933, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 9.465089797973633, | |
| "learning_rate": 8.361204013377926e-07, | |
| "loss": 1.3814, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 6.676379680633545, | |
| "learning_rate": 0.0, | |
| "loss": 1.1996, | |
| "step": 300 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 300, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 50, | |
| "total_flos": 3315779135078400.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |