| { | |
| "best_metric": 1.3153774738311768, | |
| "best_model_checkpoint": "./output/checkpoints/2024-06-11_10-50-46/checkpoint-10", | |
| "epoch": 1.0, | |
| "eval_steps": 1, | |
| "global_step": 18, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05555555555555555, | |
| "grad_norm": 1.9254094362258911, | |
| "learning_rate": 0.0002, | |
| "loss": 3.544, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.05555555555555555, | |
| "eval_loss": 3.981394052505493, | |
| "eval_runtime": 12.6022, | |
| "eval_samples_per_second": 11.347, | |
| "eval_steps_per_second": 0.714, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.1111111111111111, | |
| "grad_norm": 2.0063693523406982, | |
| "learning_rate": 0.0004, | |
| "loss": 3.6373, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.1111111111111111, | |
| "eval_loss": 3.173820972442627, | |
| "eval_runtime": 12.6737, | |
| "eval_samples_per_second": 11.283, | |
| "eval_steps_per_second": 0.71, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.16666666666666666, | |
| "grad_norm": 2.0154714584350586, | |
| "learning_rate": 0.000375, | |
| "loss": 2.7246, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.16666666666666666, | |
| "eval_loss": 1.9138473272323608, | |
| "eval_runtime": 12.6464, | |
| "eval_samples_per_second": 11.308, | |
| "eval_steps_per_second": 0.712, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.2222222222222222, | |
| "grad_norm": 1.1926450729370117, | |
| "learning_rate": 0.00035, | |
| "loss": 1.4998, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.2222222222222222, | |
| "eval_loss": 1.4986788034439087, | |
| "eval_runtime": 12.7087, | |
| "eval_samples_per_second": 11.252, | |
| "eval_steps_per_second": 0.708, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.2777777777777778, | |
| "grad_norm": 0.8283662796020508, | |
| "learning_rate": 0.00032500000000000004, | |
| "loss": 1.1373, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.2777777777777778, | |
| "eval_loss": 1.3153411149978638, | |
| "eval_runtime": 12.7034, | |
| "eval_samples_per_second": 11.257, | |
| "eval_steps_per_second": 0.708, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "grad_norm": 0.6860641241073608, | |
| "learning_rate": 0.00030000000000000003, | |
| "loss": 0.9828, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.3333333333333333, | |
| "eval_loss": 1.247497797012329, | |
| "eval_runtime": 12.6669, | |
| "eval_samples_per_second": 11.289, | |
| "eval_steps_per_second": 0.711, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.3888888888888889, | |
| "grad_norm": 0.2969834506511688, | |
| "learning_rate": 0.000275, | |
| "loss": 0.91, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.3888888888888889, | |
| "eval_loss": 1.2433477640151978, | |
| "eval_runtime": 12.6841, | |
| "eval_samples_per_second": 11.274, | |
| "eval_steps_per_second": 0.71, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 0.1941613256931305, | |
| "learning_rate": 0.00025, | |
| "loss": 0.839, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.4444444444444444, | |
| "eval_loss": 1.2696853876113892, | |
| "eval_runtime": 12.7426, | |
| "eval_samples_per_second": 11.222, | |
| "eval_steps_per_second": 0.706, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 0.19763241708278656, | |
| "learning_rate": 0.00022500000000000002, | |
| "loss": 0.7744, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "eval_loss": 1.3031294345855713, | |
| "eval_runtime": 12.7085, | |
| "eval_samples_per_second": 11.252, | |
| "eval_steps_per_second": 0.708, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.5555555555555556, | |
| "grad_norm": 0.20671840012073517, | |
| "learning_rate": 0.0002, | |
| "loss": 0.7289, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5555555555555556, | |
| "eval_loss": 1.3153774738311768, | |
| "eval_runtime": 12.6745, | |
| "eval_samples_per_second": 11.283, | |
| "eval_steps_per_second": 0.71, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.6111111111111112, | |
| "grad_norm": 0.20304171741008759, | |
| "learning_rate": 0.000175, | |
| "loss": 0.7123, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.6111111111111112, | |
| "eval_loss": 1.3239251375198364, | |
| "eval_runtime": 12.736, | |
| "eval_samples_per_second": 11.228, | |
| "eval_steps_per_second": 0.707, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "grad_norm": 0.17941166460514069, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.6761, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.6666666666666666, | |
| "eval_loss": 1.3320614099502563, | |
| "eval_runtime": 12.7074, | |
| "eval_samples_per_second": 11.253, | |
| "eval_steps_per_second": 0.708, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.7222222222222222, | |
| "grad_norm": 0.12196698784828186, | |
| "learning_rate": 0.000125, | |
| "loss": 0.6576, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.7222222222222222, | |
| "eval_loss": 1.3407816886901855, | |
| "eval_runtime": 12.7455, | |
| "eval_samples_per_second": 11.22, | |
| "eval_steps_per_second": 0.706, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.7777777777777778, | |
| "grad_norm": 0.09605628997087479, | |
| "learning_rate": 0.0001, | |
| "loss": 0.6529, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.7777777777777778, | |
| "eval_loss": 1.3476217985153198, | |
| "eval_runtime": 12.7567, | |
| "eval_samples_per_second": 11.21, | |
| "eval_steps_per_second": 0.706, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.8333333333333334, | |
| "grad_norm": 0.12586502730846405, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.6339, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.8333333333333334, | |
| "eval_loss": 1.3515379428863525, | |
| "eval_runtime": 12.6504, | |
| "eval_samples_per_second": 11.304, | |
| "eval_steps_per_second": 0.711, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 0.1299675554037094, | |
| "learning_rate": 5e-05, | |
| "loss": 0.6455, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "eval_loss": 1.3539159297943115, | |
| "eval_runtime": 12.7843, | |
| "eval_samples_per_second": 11.186, | |
| "eval_steps_per_second": 0.704, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.9444444444444444, | |
| "grad_norm": 0.1078740730881691, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.6501, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.9444444444444444, | |
| "eval_loss": 1.3558857440948486, | |
| "eval_runtime": 12.6716, | |
| "eval_samples_per_second": 11.285, | |
| "eval_steps_per_second": 0.71, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 0.15631119906902313, | |
| "learning_rate": 0.0, | |
| "loss": 0.6519, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 1.357097864151001, | |
| "eval_runtime": 12.7301, | |
| "eval_samples_per_second": 11.233, | |
| "eval_steps_per_second": 0.707, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "step": 18, | |
| "total_flos": 6363119354904576.0, | |
| "train_loss": 1.2254687315887876, | |
| "train_runtime": 391.2151, | |
| "train_samples_per_second": 1.403, | |
| "train_steps_per_second": 0.046 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 18, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 10, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6363119354904576.0, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |