| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.12175015852885225, | |
| "eval_steps": 8, | |
| "global_step": 90, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0013527795392094695, | |
| "eval_loss": 0.6261825561523438, | |
| "eval_runtime": 278.1841, | |
| "eval_samples_per_second": 8.951, | |
| "eval_steps_per_second": 4.475, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004058338617628408, | |
| "grad_norm": 0.7194657921791077, | |
| "learning_rate": 3e-05, | |
| "loss": 0.5947, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.008116677235256817, | |
| "grad_norm": 0.3807664215564728, | |
| "learning_rate": 6e-05, | |
| "loss": 0.5672, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.010822236313675756, | |
| "eval_loss": 0.5123215317726135, | |
| "eval_runtime": 277.9093, | |
| "eval_samples_per_second": 8.96, | |
| "eval_steps_per_second": 4.48, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.012175015852885225, | |
| "grad_norm": 0.30267414450645447, | |
| "learning_rate": 9e-05, | |
| "loss": 0.5331, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.016233354470513633, | |
| "grad_norm": 0.3582625389099121, | |
| "learning_rate": 0.00012, | |
| "loss": 0.4857, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.020291693088142042, | |
| "grad_norm": 0.22602972388267517, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.4491, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.021644472627351512, | |
| "eval_loss": 0.4522869288921356, | |
| "eval_runtime": 277.6814, | |
| "eval_samples_per_second": 8.967, | |
| "eval_steps_per_second": 4.484, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.02435003170577045, | |
| "grad_norm": 0.21212781965732574, | |
| "learning_rate": 0.00018, | |
| "loss": 0.4403, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.02840837032339886, | |
| "grad_norm": 0.18358972668647766, | |
| "learning_rate": 0.00019989930665413147, | |
| "loss": 0.4178, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.03246670894102727, | |
| "grad_norm": 0.19740945100784302, | |
| "learning_rate": 0.00019839295885986296, | |
| "loss": 0.4339, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.03246670894102727, | |
| "eval_loss": 0.4224604368209839, | |
| "eval_runtime": 277.593, | |
| "eval_samples_per_second": 8.97, | |
| "eval_steps_per_second": 4.485, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.036525047558655675, | |
| "grad_norm": 0.20776282250881195, | |
| "learning_rate": 0.00019510565162951537, | |
| "loss": 0.4128, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.040583386176284084, | |
| "grad_norm": 0.19233956933021545, | |
| "learning_rate": 0.0001900968867902419, | |
| "loss": 0.4056, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.043288945254703025, | |
| "eval_loss": 0.4043309986591339, | |
| "eval_runtime": 277.3816, | |
| "eval_samples_per_second": 8.977, | |
| "eval_steps_per_second": 4.488, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.04464172479391249, | |
| "grad_norm": 0.2032351940870285, | |
| "learning_rate": 0.00018345732537213027, | |
| "loss": 0.4046, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0487000634115409, | |
| "grad_norm": 0.19548796117305756, | |
| "learning_rate": 0.00017530714660036112, | |
| "loss": 0.4098, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.05275840202916931, | |
| "grad_norm": 0.19075530767440796, | |
| "learning_rate": 0.00016579387259397127, | |
| "loss": 0.39, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.054111181568378776, | |
| "eval_loss": 0.39204609394073486, | |
| "eval_runtime": 277.3667, | |
| "eval_samples_per_second": 8.977, | |
| "eval_steps_per_second": 4.489, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.05681674064679772, | |
| "grad_norm": 0.18460606038570404, | |
| "learning_rate": 0.00015508969814521025, | |
| "loss": 0.3949, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.060875079264426125, | |
| "grad_norm": 0.1991240531206131, | |
| "learning_rate": 0.00014338837391175582, | |
| "loss": 0.3922, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.06493341788205453, | |
| "grad_norm": 0.1897822767496109, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.3748, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.06493341788205453, | |
| "eval_loss": 0.38305750489234924, | |
| "eval_runtime": 277.4399, | |
| "eval_samples_per_second": 8.975, | |
| "eval_steps_per_second": 4.487, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.06899175649968295, | |
| "grad_norm": 0.18819575011730194, | |
| "learning_rate": 0.00011785568947986367, | |
| "loss": 0.3682, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.07305009511731135, | |
| "grad_norm": 0.1919279247522354, | |
| "learning_rate": 0.00010448648303505151, | |
| "loss": 0.3694, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.07575565419573028, | |
| "eval_loss": 0.3761419355869293, | |
| "eval_runtime": 277.485, | |
| "eval_samples_per_second": 8.973, | |
| "eval_steps_per_second": 4.487, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 0.07710843373493977, | |
| "grad_norm": 0.18764153122901917, | |
| "learning_rate": 9.103606910965666e-05, | |
| "loss": 0.3815, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.08116677235256817, | |
| "grad_norm": 0.18782897293567657, | |
| "learning_rate": 7.774790660436858e-05, | |
| "loss": 0.3653, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.08522511097019658, | |
| "grad_norm": 0.1975480318069458, | |
| "learning_rate": 6.486251759186572e-05, | |
| "loss": 0.3612, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.08657789050940605, | |
| "eval_loss": 0.3713741600513458, | |
| "eval_runtime": 277.5632, | |
| "eval_samples_per_second": 8.971, | |
| "eval_steps_per_second": 4.485, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 0.08928344958782498, | |
| "grad_norm": 0.15968464314937592, | |
| "learning_rate": 5.261313375270014e-05, | |
| "loss": 0.3513, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.0933417882054534, | |
| "grad_norm": 0.194020077586174, | |
| "learning_rate": 4.12214747707527e-05, | |
| "loss": 0.3362, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.0974001268230818, | |
| "grad_norm": 0.1966572403907776, | |
| "learning_rate": 3.089373510131354e-05, | |
| "loss": 0.3548, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.0974001268230818, | |
| "eval_loss": 0.3679412603378296, | |
| "eval_runtime": 277.5408, | |
| "eval_samples_per_second": 8.972, | |
| "eval_steps_per_second": 4.486, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.10145846544071022, | |
| "grad_norm": 0.1846090853214264, | |
| "learning_rate": 2.181685175319702e-05, | |
| "loss": 0.356, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.10551680405833862, | |
| "grad_norm": 0.18600720167160034, | |
| "learning_rate": 1.415512063981339e-05, | |
| "loss": 0.3634, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.10822236313675755, | |
| "eval_loss": 0.36580997705459595, | |
| "eval_runtime": 277.4172, | |
| "eval_samples_per_second": 8.976, | |
| "eval_steps_per_second": 4.488, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.10957514267596703, | |
| "grad_norm": 0.18399472534656525, | |
| "learning_rate": 8.047222744854943e-06, | |
| "loss": 0.3714, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.11363348129359543, | |
| "grad_norm": 0.1955898106098175, | |
| "learning_rate": 3.6037139304146762e-06, | |
| "loss": 0.3674, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.11769181991122385, | |
| "grad_norm": 0.1757526993751526, | |
| "learning_rate": 9.0502382320653e-07, | |
| "loss": 0.3592, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.11904459945043332, | |
| "eval_loss": 0.3652554452419281, | |
| "eval_runtime": 277.5208, | |
| "eval_samples_per_second": 8.972, | |
| "eval_steps_per_second": 4.486, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 0.12175015852885225, | |
| "grad_norm": 0.17539910972118378, | |
| "learning_rate": 0.0, | |
| "loss": 0.3584, | |
| "step": 90 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 90, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 8, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.373140281715589e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |