| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.10970927043335162, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0021941854086670325, | |
| "grad_norm": NaN, | |
| "learning_rate": 0.0, | |
| "loss": 9.7459, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0021941854086670325, | |
| "eval_loss": 9.789505004882812, | |
| "eval_runtime": 23.1683, | |
| "eval_samples_per_second": 8.287, | |
| "eval_steps_per_second": 4.144, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.004388370817334065, | |
| "grad_norm": 17.736719131469727, | |
| "learning_rate": 1e-05, | |
| "loss": 10.0567, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.006582556226001097, | |
| "grad_norm": 18.50017738342285, | |
| "learning_rate": 2e-05, | |
| "loss": 9.2557, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00877674163466813, | |
| "grad_norm": 17.195247650146484, | |
| "learning_rate": 3e-05, | |
| "loss": 8.6156, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.010970927043335162, | |
| "grad_norm": 18.478811264038086, | |
| "learning_rate": 4e-05, | |
| "loss": 8.6089, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.010970927043335162, | |
| "eval_loss": 8.13350772857666, | |
| "eval_runtime": 22.7364, | |
| "eval_samples_per_second": 8.445, | |
| "eval_steps_per_second": 4.222, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.013165112452002194, | |
| "grad_norm": 14.635212898254395, | |
| "learning_rate": 5e-05, | |
| "loss": 8.6232, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.015359297860669226, | |
| "grad_norm": 13.62023639678955, | |
| "learning_rate": 6e-05, | |
| "loss": 6.725, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.01755348326933626, | |
| "grad_norm": 14.408036231994629, | |
| "learning_rate": 7e-05, | |
| "loss": 5.464, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.019747668678003292, | |
| "grad_norm": 17.330280303955078, | |
| "learning_rate": 8e-05, | |
| "loss": 4.2439, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.021941854086670324, | |
| "grad_norm": 15.395853996276855, | |
| "learning_rate": 9e-05, | |
| "loss": 2.9754, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.021941854086670324, | |
| "eval_loss": 1.9544070959091187, | |
| "eval_runtime": 22.7951, | |
| "eval_samples_per_second": 8.423, | |
| "eval_steps_per_second": 4.211, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.024136039495337356, | |
| "grad_norm": 14.372428894042969, | |
| "learning_rate": 0.0001, | |
| "loss": 2.1836, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.026330224904004388, | |
| "grad_norm": 10.696700096130371, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 0.8312, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.02852441031267142, | |
| "grad_norm": 6.342767238616943, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 0.6199, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.030718595721338452, | |
| "grad_norm": 8.464292526245117, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 0.5169, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.03291278113000549, | |
| "grad_norm": 7.956654071807861, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.765, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.03291278113000549, | |
| "eval_loss": 0.5048862099647522, | |
| "eval_runtime": 22.8072, | |
| "eval_samples_per_second": 8.418, | |
| "eval_steps_per_second": 4.209, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.03510696653867252, | |
| "grad_norm": 10.383864402770996, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 0.7324, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.03730115194733955, | |
| "grad_norm": 4.3315110206604, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 0.4558, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.039495337356006584, | |
| "grad_norm": 19.855714797973633, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 1.2013, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.041689522764673616, | |
| "grad_norm": 8.671701431274414, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 0.8788, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.04388370817334065, | |
| "grad_norm": 8.814596176147461, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 0.9201, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.04388370817334065, | |
| "eval_loss": 0.7541892528533936, | |
| "eval_runtime": 22.8305, | |
| "eval_samples_per_second": 8.41, | |
| "eval_steps_per_second": 4.205, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.04607789358200768, | |
| "grad_norm": 1.8071426153182983, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 0.4432, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.04827207899067471, | |
| "grad_norm": 4.000314235687256, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 0.6024, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.050466264399341744, | |
| "grad_norm": 4.672601222991943, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 0.494, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.052660449808008776, | |
| "grad_norm": 2.561952590942383, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 0.4829, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.05485463521667581, | |
| "grad_norm": 5.277298927307129, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 0.5405, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.05485463521667581, | |
| "eval_loss": 0.5149335861206055, | |
| "eval_runtime": 22.8748, | |
| "eval_samples_per_second": 8.394, | |
| "eval_steps_per_second": 4.197, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.05704882062534284, | |
| "grad_norm": 1.4059255123138428, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 0.3548, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.05924300603400987, | |
| "grad_norm": 3.23660945892334, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 0.5226, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.061437191442676904, | |
| "grad_norm": 9.492186546325684, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 0.4232, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.06363137685134394, | |
| "grad_norm": 1.2060160636901855, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 0.323, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.06582556226001098, | |
| "grad_norm": 3.9120266437530518, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 0.5533, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.06582556226001098, | |
| "eval_loss": 0.5817310810089111, | |
| "eval_runtime": 22.8537, | |
| "eval_samples_per_second": 8.401, | |
| "eval_steps_per_second": 4.201, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.06801974766867801, | |
| "grad_norm": 8.79382038116455, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0141, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.07021393307734504, | |
| "grad_norm": 7.010884761810303, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 0.4862, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.07240811848601207, | |
| "grad_norm": 10.745322227478027, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 1.0003, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0746023038946791, | |
| "grad_norm": 3.4400458335876465, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 0.3171, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.07679648930334614, | |
| "grad_norm": 12.523181915283203, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 0.8103, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.07679648930334614, | |
| "eval_loss": 0.4702953100204468, | |
| "eval_runtime": 22.7974, | |
| "eval_samples_per_second": 8.422, | |
| "eval_steps_per_second": 4.211, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.07899067471201317, | |
| "grad_norm": 10.607621192932129, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 0.7069, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0811848601206802, | |
| "grad_norm": 5.708343982696533, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 0.3979, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.08337904552934723, | |
| "grad_norm": 7.0952348709106445, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 0.5199, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.08557323093801426, | |
| "grad_norm": 6.050652980804443, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 0.4952, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0877674163466813, | |
| "grad_norm": 3.988201141357422, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 0.4951, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0877674163466813, | |
| "eval_loss": 0.4913504123687744, | |
| "eval_runtime": 22.8247, | |
| "eval_samples_per_second": 8.412, | |
| "eval_steps_per_second": 4.206, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.08996160175534833, | |
| "grad_norm": 3.5365755558013916, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 0.4151, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.09215578716401536, | |
| "grad_norm": 2.981470823287964, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 0.3609, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.09434997257268239, | |
| "grad_norm": 7.036005020141602, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 0.4929, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.09654415798134942, | |
| "grad_norm": 16.351613998413086, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 0.8053, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.09873834339001646, | |
| "grad_norm": 7.42625093460083, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 0.4763, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.09873834339001646, | |
| "eval_loss": 0.44910240173339844, | |
| "eval_runtime": 22.8479, | |
| "eval_samples_per_second": 8.403, | |
| "eval_steps_per_second": 4.202, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.10093252879868349, | |
| "grad_norm": 6.476613521575928, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 0.4703, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.10312671420735052, | |
| "grad_norm": 6.421245098114014, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 0.4754, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.10532089961601755, | |
| "grad_norm": 3.1271674633026123, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 0.3527, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.10751508502468458, | |
| "grad_norm": 6.350700855255127, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 0.4567, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.10970927043335162, | |
| "grad_norm": 2.4120147228240967, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 0.3867, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.10970927043335162, | |
| "eval_loss": 0.41695770621299744, | |
| "eval_runtime": 22.8514, | |
| "eval_samples_per_second": 8.402, | |
| "eval_steps_per_second": 4.201, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.85471820890112e+16, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |