| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.0, | |
| "eval_steps": 500, | |
| "global_step": 5000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 5.201437473297119, | |
| "learning_rate": 4.9e-05, | |
| "loss": 2.6116, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 4.178571701049805, | |
| "learning_rate": 4.8e-05, | |
| "loss": 1.9634, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 3.5103917121887207, | |
| "learning_rate": 4.7e-05, | |
| "loss": 1.6752, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 4.7221550941467285, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 1.483, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 4.4923272132873535, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.3296, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 6.077773571014404, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 1.1978, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 2.388939142227173, | |
| "learning_rate": 4.3e-05, | |
| "loss": 1.1306, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 2.460210084915161, | |
| "learning_rate": 4.201e-05, | |
| "loss": 1.0513, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 2.139218330383301, | |
| "learning_rate": 4.101e-05, | |
| "loss": 0.9612, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 2.382781505584717, | |
| "learning_rate": 4.0010000000000005e-05, | |
| "loss": 0.8977, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 6.498015403747559, | |
| "learning_rate": 3.902e-05, | |
| "loss": 0.8653, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 2.856051445007324, | |
| "learning_rate": 3.802e-05, | |
| "loss": 0.8131, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 2.134838819503784, | |
| "learning_rate": 3.702e-05, | |
| "loss": 0.7421, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 2.3750529289245605, | |
| "learning_rate": 3.6020000000000004e-05, | |
| "loss": 0.6864, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 2.56854510307312, | |
| "learning_rate": 3.502e-05, | |
| "loss": 0.6453, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 2.0850515365600586, | |
| "learning_rate": 3.402e-05, | |
| "loss": 0.6245, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "grad_norm": 2.8658158779144287, | |
| "learning_rate": 3.302e-05, | |
| "loss": 0.597, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 2.8391458988189697, | |
| "learning_rate": 3.202e-05, | |
| "loss": 0.5792, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 2.1407833099365234, | |
| "learning_rate": 3.102e-05, | |
| "loss": 0.5501, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 3.772876739501953, | |
| "learning_rate": 3.0020000000000004e-05, | |
| "loss": 0.5411, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "grad_norm": 1.72440505027771, | |
| "learning_rate": 2.9020000000000003e-05, | |
| "loss": 0.5107, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 3.845543622970581, | |
| "learning_rate": 2.8020000000000003e-05, | |
| "loss": 0.4988, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 1.5010182857513428, | |
| "learning_rate": 2.7020000000000002e-05, | |
| "loss": 0.48, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 1.9300732612609863, | |
| "learning_rate": 2.602e-05, | |
| "loss": 0.4683, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.938793659210205, | |
| "learning_rate": 2.5019999999999998e-05, | |
| "loss": 0.4501, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 1.3520616292953491, | |
| "learning_rate": 2.402e-05, | |
| "loss": 0.3959, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 2.16, | |
| "grad_norm": 1.9204097986221313, | |
| "learning_rate": 2.302e-05, | |
| "loss": 0.3846, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 2.24, | |
| "grad_norm": 2.809589385986328, | |
| "learning_rate": 2.2020000000000003e-05, | |
| "loss": 0.3766, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 2.32, | |
| "grad_norm": 2.0556161403656006, | |
| "learning_rate": 2.1020000000000002e-05, | |
| "loss": 0.3681, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 1.61994206905365, | |
| "learning_rate": 2.002e-05, | |
| "loss": 0.3599, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 2.48, | |
| "grad_norm": 2.605128526687622, | |
| "learning_rate": 1.902e-05, | |
| "loss": 0.3499, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "grad_norm": 1.2965134382247925, | |
| "learning_rate": 1.802e-05, | |
| "loss": 0.3351, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 2.64, | |
| "grad_norm": 1.3328615427017212, | |
| "learning_rate": 1.702e-05, | |
| "loss": 0.3404, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 2.72, | |
| "grad_norm": 1.6524440050125122, | |
| "learning_rate": 1.6020000000000002e-05, | |
| "loss": 0.3357, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 1.3374449014663696, | |
| "learning_rate": 1.502e-05, | |
| "loss": 0.321, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "grad_norm": 1.1381638050079346, | |
| "learning_rate": 1.402e-05, | |
| "loss": 0.3157, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "grad_norm": 1.3332613706588745, | |
| "learning_rate": 1.3020000000000002e-05, | |
| "loss": 0.3069, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 3.04, | |
| "grad_norm": 1.121593713760376, | |
| "learning_rate": 1.202e-05, | |
| "loss": 0.2887, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 3.12, | |
| "grad_norm": 1.3822081089019775, | |
| "learning_rate": 1.1020000000000001e-05, | |
| "loss": 0.2715, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 1.5630627870559692, | |
| "learning_rate": 1.002e-05, | |
| "loss": 0.2685, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 3.28, | |
| "grad_norm": 1.2118648290634155, | |
| "learning_rate": 9.02e-06, | |
| "loss": 0.2669, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 3.36, | |
| "grad_norm": 1.0490390062332153, | |
| "learning_rate": 8.02e-06, | |
| "loss": 0.2545, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 3.44, | |
| "grad_norm": 1.6597875356674194, | |
| "learning_rate": 7.0200000000000006e-06, | |
| "loss": 0.2599, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 3.52, | |
| "grad_norm": 1.3064651489257812, | |
| "learning_rate": 6.02e-06, | |
| "loss": 0.2599, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 1.6653597354888916, | |
| "learning_rate": 5.02e-06, | |
| "loss": 0.2561, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "grad_norm": 1.1606736183166504, | |
| "learning_rate": 4.0200000000000005e-06, | |
| "loss": 0.2481, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 3.76, | |
| "grad_norm": 0.9921647310256958, | |
| "learning_rate": 3.0200000000000003e-06, | |
| "loss": 0.2481, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 3.84, | |
| "grad_norm": 1.1727889776229858, | |
| "learning_rate": 2.02e-06, | |
| "loss": 0.2433, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 3.92, | |
| "grad_norm": 1.4723715782165527, | |
| "learning_rate": 1.0200000000000002e-06, | |
| "loss": 0.2427, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 1.1171002388000488, | |
| "learning_rate": 2e-08, | |
| "loss": 0.2419, | |
| "step": 5000 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 5000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 500, | |
| "total_flos": 7.671177692577792e+20, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |