| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.23668639053254437, | |
| "eval_steps": 20, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.001183431952662722, | |
| "eval_loss": 6.91423225402832, | |
| "eval_runtime": 2.3165, | |
| "eval_samples_per_second": 648.398, | |
| "eval_steps_per_second": 40.579, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.011834319526627219, | |
| "grad_norm": 8640.0, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 6.7118, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.023668639053254437, | |
| "grad_norm": 2368.0, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 6.7944, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.023668639053254437, | |
| "eval_loss": 6.8260087966918945, | |
| "eval_runtime": 2.3051, | |
| "eval_samples_per_second": 651.607, | |
| "eval_steps_per_second": 40.78, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.03550295857988166, | |
| "grad_norm": 8096.0, | |
| "learning_rate": 4.8e-05, | |
| "loss": 6.762, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.047337278106508875, | |
| "grad_norm": 4160.0, | |
| "learning_rate": 6.400000000000001e-05, | |
| "loss": 6.5961, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.047337278106508875, | |
| "eval_loss": 6.617987632751465, | |
| "eval_runtime": 2.3313, | |
| "eval_samples_per_second": 644.27, | |
| "eval_steps_per_second": 40.321, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05917159763313609, | |
| "grad_norm": 17920.0, | |
| "learning_rate": 8e-05, | |
| "loss": 6.4182, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.07100591715976332, | |
| "grad_norm": 13632.0, | |
| "learning_rate": 9.6e-05, | |
| "loss": 6.5851, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.07100591715976332, | |
| "eval_loss": 6.849647045135498, | |
| "eval_runtime": 2.5266, | |
| "eval_samples_per_second": 594.476, | |
| "eval_steps_per_second": 37.204, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.08284023668639054, | |
| "grad_norm": 11136.0, | |
| "learning_rate": 0.00011200000000000001, | |
| "loss": 7.1215, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.09467455621301775, | |
| "grad_norm": 13568.0, | |
| "learning_rate": 0.00012800000000000002, | |
| "loss": 6.9277, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.09467455621301775, | |
| "eval_loss": 7.1117143630981445, | |
| "eval_runtime": 2.3547, | |
| "eval_samples_per_second": 637.865, | |
| "eval_steps_per_second": 39.92, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.10650887573964497, | |
| "grad_norm": 7520.0, | |
| "learning_rate": 0.000144, | |
| "loss": 6.9477, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.11834319526627218, | |
| "grad_norm": 5344.0, | |
| "learning_rate": 0.00016, | |
| "loss": 7.4897, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.11834319526627218, | |
| "eval_loss": 7.819643497467041, | |
| "eval_runtime": 2.3681, | |
| "eval_samples_per_second": 634.257, | |
| "eval_steps_per_second": 39.694, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.1301775147928994, | |
| "grad_norm": 1944.0, | |
| "learning_rate": 0.00017600000000000002, | |
| "loss": 8.0353, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.14201183431952663, | |
| "grad_norm": 386.0, | |
| "learning_rate": 0.000192, | |
| "loss": 8.3949, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.14201183431952663, | |
| "eval_loss": 8.92322063446045, | |
| "eval_runtime": 2.2886, | |
| "eval_samples_per_second": 656.309, | |
| "eval_steps_per_second": 41.074, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.15384615384615385, | |
| "grad_norm": 2752.0, | |
| "learning_rate": 0.0001999978128380225, | |
| "loss": 8.8096, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.16568047337278108, | |
| "grad_norm": 2208.0, | |
| "learning_rate": 0.0001999803161162393, | |
| "loss": 9.3893, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.16568047337278108, | |
| "eval_loss": 10.95390510559082, | |
| "eval_runtime": 2.3845, | |
| "eval_samples_per_second": 629.893, | |
| "eval_steps_per_second": 39.421, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.17751479289940827, | |
| "grad_norm": 1104.0, | |
| "learning_rate": 0.00019994532573409262, | |
| "loss": 10.2318, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.1893491124260355, | |
| "grad_norm": 664.0, | |
| "learning_rate": 0.00019989284781388617, | |
| "loss": 8.9254, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.1893491124260355, | |
| "eval_loss": 9.343338966369629, | |
| "eval_runtime": 2.3475, | |
| "eval_samples_per_second": 639.824, | |
| "eval_steps_per_second": 40.042, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.20118343195266272, | |
| "grad_norm": 576.0, | |
| "learning_rate": 0.00019982289153773646, | |
| "loss": 8.9456, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.21301775147928995, | |
| "grad_norm": 3888.0, | |
| "learning_rate": 0.00019973546914596623, | |
| "loss": 9.1233, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.21301775147928995, | |
| "eval_loss": 9.207473754882812, | |
| "eval_runtime": 2.8413, | |
| "eval_samples_per_second": 528.635, | |
| "eval_steps_per_second": 33.084, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.22485207100591717, | |
| "grad_norm": 884.0, | |
| "learning_rate": 0.00019963059593496268, | |
| "loss": 9.3116, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.23668639053254437, | |
| "grad_norm": 410.0, | |
| "learning_rate": 0.00019950829025450114, | |
| "loss": 9.2642, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.23668639053254437, | |
| "eval_loss": 9.775616645812988, | |
| "eval_runtime": 2.3679, | |
| "eval_samples_per_second": 634.329, | |
| "eval_steps_per_second": 39.698, | |
| "step": 200 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 2500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 40, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 300220566994944.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |