| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 1.556420233463035, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01556420233463035, | |
| "eval_loss": 0.888220489025116, | |
| "eval_runtime": 7.1127, | |
| "eval_samples_per_second": 15.184, | |
| "eval_steps_per_second": 1.968, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.04669260700389105, | |
| "grad_norm": 3.0385119915008545, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.863, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0933852140077821, | |
| "grad_norm": 2.6818928718566895, | |
| "learning_rate": 3e-05, | |
| "loss": 0.8733, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.14007782101167315, | |
| "grad_norm": 0.7148049473762512, | |
| "learning_rate": 4.5e-05, | |
| "loss": 0.8218, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.14007782101167315, | |
| "eval_loss": 0.7536948919296265, | |
| "eval_runtime": 7.0144, | |
| "eval_samples_per_second": 15.397, | |
| "eval_steps_per_second": 1.996, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.1867704280155642, | |
| "grad_norm": 0.43349385261535645, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 0.7216, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.23346303501945526, | |
| "grad_norm": 0.3839814066886902, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 0.722, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.2801556420233463, | |
| "grad_norm": 0.40558719635009766, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 0.6981, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.2801556420233463, | |
| "eval_loss": 0.6632740497589111, | |
| "eval_runtime": 7.0535, | |
| "eval_samples_per_second": 15.312, | |
| "eval_steps_per_second": 1.985, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.32684824902723736, | |
| "grad_norm": 0.35617005825042725, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 0.6165, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.3735408560311284, | |
| "grad_norm": 0.3675176799297333, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 0.699, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.42023346303501946, | |
| "grad_norm": 0.32696929574012756, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 0.6745, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.42023346303501946, | |
| "eval_loss": 0.6363734006881714, | |
| "eval_runtime": 7.072, | |
| "eval_samples_per_second": 15.271, | |
| "eval_steps_per_second": 1.98, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.4669260700389105, | |
| "grad_norm": 0.28614386916160583, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 0.633, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.5136186770428015, | |
| "grad_norm": 0.32294508814811707, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 0.6074, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.5603112840466926, | |
| "grad_norm": 0.3273395895957947, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 0.6574, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.5603112840466926, | |
| "eval_loss": 0.6291098594665527, | |
| "eval_runtime": 7.0767, | |
| "eval_samples_per_second": 15.261, | |
| "eval_steps_per_second": 1.978, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.6070038910505836, | |
| "grad_norm": 0.27768364548683167, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 0.639, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.6536964980544747, | |
| "grad_norm": 0.29399535059928894, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 0.6505, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.7003891050583657, | |
| "grad_norm": 0.3063369393348694, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 0.6367, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.7003891050583657, | |
| "eval_loss": 0.6239914298057556, | |
| "eval_runtime": 7.0733, | |
| "eval_samples_per_second": 15.269, | |
| "eval_steps_per_second": 1.979, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.7470817120622568, | |
| "grad_norm": 0.3164641559123993, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 0.7004, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.7937743190661478, | |
| "grad_norm": 0.29213565587997437, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 0.6424, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.8404669260700389, | |
| "grad_norm": 0.31585273146629333, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 0.6445, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.8404669260700389, | |
| "eval_loss": 0.6201390624046326, | |
| "eval_runtime": 7.08, | |
| "eval_samples_per_second": 15.254, | |
| "eval_steps_per_second": 1.977, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.8871595330739299, | |
| "grad_norm": 0.31377464532852173, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 0.6538, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.933852140077821, | |
| "grad_norm": 0.2973349988460541, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 0.6141, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.980544747081712, | |
| "grad_norm": 0.31598252058029175, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 0.6287, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.980544747081712, | |
| "eval_loss": 0.6172313094139099, | |
| "eval_runtime": 7.2626, | |
| "eval_samples_per_second": 14.871, | |
| "eval_steps_per_second": 1.928, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 1.027237354085603, | |
| "grad_norm": 0.2909192144870758, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 0.776, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 1.0739299610894941, | |
| "grad_norm": 0.2928639054298401, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 0.6068, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 1.1206225680933852, | |
| "grad_norm": 0.2987300753593445, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 0.5927, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.1206225680933852, | |
| "eval_loss": 0.6153312921524048, | |
| "eval_runtime": 7.261, | |
| "eval_samples_per_second": 14.874, | |
| "eval_steps_per_second": 1.928, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 1.1673151750972763, | |
| "grad_norm": 0.3104216456413269, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 0.6262, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.2140077821011672, | |
| "grad_norm": 0.2893487215042114, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 0.623, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 1.2607003891050583, | |
| "grad_norm": 0.309826523065567, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 0.643, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.2607003891050583, | |
| "eval_loss": 0.614518404006958, | |
| "eval_runtime": 7.0859, | |
| "eval_samples_per_second": 15.242, | |
| "eval_steps_per_second": 1.976, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 1.3073929961089494, | |
| "grad_norm": 0.3125797212123871, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 0.5811, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 1.3540856031128405, | |
| "grad_norm": 0.31804636120796204, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 0.6013, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 1.4007782101167314, | |
| "grad_norm": 0.31651419401168823, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 0.6143, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.4007782101167314, | |
| "eval_loss": 0.6147182583808899, | |
| "eval_runtime": 7.2685, | |
| "eval_samples_per_second": 14.859, | |
| "eval_steps_per_second": 1.926, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.4474708171206225, | |
| "grad_norm": 0.2994842529296875, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 0.5747, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 1.4941634241245136, | |
| "grad_norm": 0.3009912073612213, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 0.6074, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 1.5408560311284045, | |
| "grad_norm": 0.3032476603984833, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 0.6417, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 1.5408560311284045, | |
| "eval_loss": 0.6142562031745911, | |
| "eval_runtime": 7.0785, | |
| "eval_samples_per_second": 15.257, | |
| "eval_steps_per_second": 1.978, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 7.220911035069235e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |