| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 10.0, | |
| "eval_steps": 1000.0, | |
| "global_step": 25000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 3.1640737056732178, | |
| "learning_rate": 4.9e-05, | |
| "loss": 2.212, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 23.492929458618164, | |
| "learning_rate": 4.8e-05, | |
| "loss": 2.0008, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 2.86879563331604, | |
| "learning_rate": 4.7e-05, | |
| "loss": 1.9454, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 2.5470571517944336, | |
| "learning_rate": 4.600000000000001e-05, | |
| "loss": 1.9098, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 3.299417018890381, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.8775, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 3.0483977794647217, | |
| "learning_rate": 4.4000000000000006e-05, | |
| "loss": 1.8398, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 3.0028903484344482, | |
| "learning_rate": 4.3e-05, | |
| "loss": 1.8147, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 2.972844123840332, | |
| "learning_rate": 4.2e-05, | |
| "loss": 1.8035, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 2.82574200630188, | |
| "learning_rate": 4.1e-05, | |
| "loss": 1.7974, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.6531100273132324, | |
| "learning_rate": 4e-05, | |
| "loss": 1.7818, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.2, | |
| "grad_norm": 3.6078109741210938, | |
| "learning_rate": 3.9000000000000006e-05, | |
| "loss": 1.7492, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.4, | |
| "grad_norm": 2.573765516281128, | |
| "learning_rate": 3.8e-05, | |
| "loss": 1.7536, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.6, | |
| "grad_norm": 3.07545804977417, | |
| "learning_rate": 3.7e-05, | |
| "loss": 1.7341, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.8, | |
| "grad_norm": 2.4871368408203125, | |
| "learning_rate": 3.6e-05, | |
| "loss": 1.7362, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "grad_norm": 2.881721019744873, | |
| "learning_rate": 3.5e-05, | |
| "loss": 1.7426, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "grad_norm": 3.6232352256774902, | |
| "learning_rate": 3.4000000000000007e-05, | |
| "loss": 1.6958, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "grad_norm": 2.4172329902648926, | |
| "learning_rate": 3.3e-05, | |
| "loss": 1.6961, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.6, | |
| "grad_norm": 2.5114290714263916, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 1.6961, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.8, | |
| "grad_norm": 2.7974891662597656, | |
| "learning_rate": 3.1e-05, | |
| "loss": 1.6831, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 2.511326551437378, | |
| "learning_rate": 3e-05, | |
| "loss": 1.7117, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 4.2, | |
| "grad_norm": 2.3756415843963623, | |
| "learning_rate": 2.9e-05, | |
| "loss": 1.6684, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 4.4, | |
| "grad_norm": 2.2897322177886963, | |
| "learning_rate": 2.8000000000000003e-05, | |
| "loss": 1.6826, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 4.6, | |
| "grad_norm": 2.7029635906219482, | |
| "learning_rate": 2.7000000000000002e-05, | |
| "loss": 1.6649, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 2.453059673309326, | |
| "learning_rate": 2.6000000000000002e-05, | |
| "loss": 1.6488, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 2.173687219619751, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.652, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 5.2, | |
| "grad_norm": 2.9122655391693115, | |
| "learning_rate": 2.4e-05, | |
| "loss": 1.6354, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 5.4, | |
| "grad_norm": 2.8792638778686523, | |
| "learning_rate": 2.3000000000000003e-05, | |
| "loss": 1.6423, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 5.6, | |
| "grad_norm": 3.0389626026153564, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 1.6591, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 5.8, | |
| "grad_norm": 2.6412346363067627, | |
| "learning_rate": 2.1e-05, | |
| "loss": 1.6365, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "grad_norm": 2.6655163764953613, | |
| "learning_rate": 2e-05, | |
| "loss": 1.6133, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 6.2, | |
| "grad_norm": 2.6776795387268066, | |
| "learning_rate": 1.9e-05, | |
| "loss": 1.603, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "grad_norm": 3.4206550121307373, | |
| "learning_rate": 1.8e-05, | |
| "loss": 1.6142, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 6.6, | |
| "grad_norm": 3.59487247467041, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 1.6193, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "grad_norm": 2.707260847091675, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.618, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "grad_norm": 2.4932329654693604, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.6179, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 7.2, | |
| "grad_norm": 2.7114312648773193, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 1.6064, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 7.4, | |
| "grad_norm": 2.9900901317596436, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 1.6079, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 7.6, | |
| "grad_norm": 2.5069007873535156, | |
| "learning_rate": 1.2e-05, | |
| "loss": 1.5917, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "grad_norm": 2.816830635070801, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 1.6159, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 2.500258445739746, | |
| "learning_rate": 1e-05, | |
| "loss": 1.586, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 8.2, | |
| "grad_norm": 2.392134189605713, | |
| "learning_rate": 9e-06, | |
| "loss": 1.5948, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "grad_norm": 2.19275164604187, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.5904, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 8.6, | |
| "grad_norm": 2.1356213092803955, | |
| "learning_rate": 7.000000000000001e-06, | |
| "loss": 1.5772, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 8.8, | |
| "grad_norm": 2.0939230918884277, | |
| "learning_rate": 6e-06, | |
| "loss": 1.604, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "grad_norm": 2.8503177165985107, | |
| "learning_rate": 5e-06, | |
| "loss": 1.5878, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 9.2, | |
| "grad_norm": 2.6212220191955566, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.5893, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 9.4, | |
| "grad_norm": 2.9165279865264893, | |
| "learning_rate": 3e-06, | |
| "loss": 1.5796, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "grad_norm": 2.8491413593292236, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 1.5832, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 9.8, | |
| "grad_norm": 3.690595865249634, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 1.5872, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 3.193439245223999, | |
| "learning_rate": 0.0, | |
| "loss": 1.5781, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 25000, | |
| "total_flos": 3.2973681787109376e+16, | |
| "train_loss": 1.6927287036132812, | |
| "train_runtime": 2178.3738, | |
| "train_samples_per_second": 91.812, | |
| "train_steps_per_second": 11.476 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 25000, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.2973681787109376e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |