| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.9790209790209791, | |
| "eval_steps": 9, | |
| "global_step": 35, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.027972027972027972, | |
| "grad_norm": 5.75, | |
| "learning_rate": 2e-05, | |
| "loss": 1.8618, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.027972027972027972, | |
| "eval_loss": 1.856866478919983, | |
| "eval_runtime": 4.5304, | |
| "eval_samples_per_second": 3.532, | |
| "eval_steps_per_second": 1.766, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.055944055944055944, | |
| "grad_norm": 5.5, | |
| "learning_rate": 4e-05, | |
| "loss": 1.8343, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.08391608391608392, | |
| "grad_norm": 5.75, | |
| "learning_rate": 6e-05, | |
| "loss": 1.87, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.11188811188811189, | |
| "grad_norm": 6.0625, | |
| "learning_rate": 8e-05, | |
| "loss": 1.6976, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.13986013986013987, | |
| "grad_norm": 5.6875, | |
| "learning_rate": 0.0001, | |
| "loss": 1.0745, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.16783216783216784, | |
| "grad_norm": 2.875, | |
| "learning_rate": 0.00012, | |
| "loss": 0.5387, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.1958041958041958, | |
| "grad_norm": 2.90625, | |
| "learning_rate": 0.00014, | |
| "loss": 0.3055, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.22377622377622378, | |
| "grad_norm": 0.51171875, | |
| "learning_rate": 0.00016, | |
| "loss": 0.0592, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.2517482517482518, | |
| "grad_norm": 0.65625, | |
| "learning_rate": 0.00018, | |
| "loss": 0.0185, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.2517482517482518, | |
| "eval_loss": 0.0595741793513298, | |
| "eval_runtime": 4.6321, | |
| "eval_samples_per_second": 3.454, | |
| "eval_steps_per_second": 1.727, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.27972027972027974, | |
| "grad_norm": 1.65625, | |
| "learning_rate": 0.0002, | |
| "loss": 0.1101, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.3076923076923077, | |
| "grad_norm": 6.375, | |
| "learning_rate": 0.00019997080140801932, | |
| "loss": 0.1327, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.3356643356643357, | |
| "grad_norm": 0.21484375, | |
| "learning_rate": 0.00019988322268323268, | |
| "loss": 0.0082, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 0.1357421875, | |
| "learning_rate": 0.00019973731496914914, | |
| "loss": 0.0049, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.3916083916083916, | |
| "grad_norm": 0.0693359375, | |
| "learning_rate": 0.00019953316347176488, | |
| "loss": 0.0024, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.4195804195804196, | |
| "grad_norm": 0.49609375, | |
| "learning_rate": 0.0001992708874098054, | |
| "loss": 0.065, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.44755244755244755, | |
| "grad_norm": 0.07568359375, | |
| "learning_rate": 0.0001989506399451051, | |
| "loss": 0.0029, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.4755244755244755, | |
| "grad_norm": 0.134765625, | |
| "learning_rate": 0.0001985726080931651, | |
| "loss": 0.0053, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.5034965034965035, | |
| "grad_norm": 0.1416015625, | |
| "learning_rate": 0.00019813701261394136, | |
| "loss": 0.0056, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.5034965034965035, | |
| "eval_loss": 0.02016551047563553, | |
| "eval_runtime": 4.6628, | |
| "eval_samples_per_second": 3.431, | |
| "eval_steps_per_second": 1.716, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.5314685314685315, | |
| "grad_norm": 0.0791015625, | |
| "learning_rate": 0.00019764410788292722, | |
| "loss": 0.003, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.5594405594405595, | |
| "grad_norm": 0.032470703125, | |
| "learning_rate": 0.0001970941817426052, | |
| "loss": 0.0011, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.5874125874125874, | |
| "grad_norm": 0.447265625, | |
| "learning_rate": 0.00019648755533435518, | |
| "loss": 0.0445, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.6153846153846154, | |
| "grad_norm": 0.87890625, | |
| "learning_rate": 0.00019582458291091663, | |
| "loss": 0.0708, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.6433566433566433, | |
| "grad_norm": 0.259765625, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.0114, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.6713286713286714, | |
| "grad_norm": 0.609375, | |
| "learning_rate": 0.0001943311813257743, | |
| "loss": 0.0243, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.6993006993006993, | |
| "grad_norm": 1.921875, | |
| "learning_rate": 0.0001935016242685415, | |
| "loss": 0.0655, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.515625, | |
| "learning_rate": 0.00019261746489577765, | |
| "loss": 0.0756, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.7552447552447552, | |
| "grad_norm": 0.0201416015625, | |
| "learning_rate": 0.00019167921953165825, | |
| "loss": 0.0008, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.7552447552447552, | |
| "eval_loss": 0.0004657781682908535, | |
| "eval_runtime": 4.6398, | |
| "eval_samples_per_second": 3.448, | |
| "eval_steps_per_second": 1.724, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.7832167832167832, | |
| "grad_norm": 0.0186767578125, | |
| "learning_rate": 0.00019068743608505455, | |
| "loss": 0.0008, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.8111888111888111, | |
| "grad_norm": 0.02392578125, | |
| "learning_rate": 0.00018964269372957038, | |
| "loss": 0.0008, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.8391608391608392, | |
| "grad_norm": 0.134765625, | |
| "learning_rate": 0.000188545602565321, | |
| "loss": 0.0046, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.8671328671328671, | |
| "grad_norm": 0.1416015625, | |
| "learning_rate": 0.0001873968032626518, | |
| "loss": 0.0049, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.8951048951048951, | |
| "grad_norm": 0.1220703125, | |
| "learning_rate": 0.00018619696668800492, | |
| "loss": 0.0047, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.9230769230769231, | |
| "grad_norm": 0.0240478515625, | |
| "learning_rate": 0.0001849467935121521, | |
| "loss": 0.0012, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.951048951048951, | |
| "grad_norm": 0.032958984375, | |
| "learning_rate": 0.00018364701380102266, | |
| "loss": 0.0014, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.9790209790209791, | |
| "grad_norm": 0.0247802734375, | |
| "learning_rate": 0.00018229838658936564, | |
| "loss": 0.001, | |
| "step": 35 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 140, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 35, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.222073165152256e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |