| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.0015423768026528882, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 3.084753605305776e-05, | |
| "grad_norm": 20.13319206237793, | |
| "learning_rate": 1e-05, | |
| "loss": 6.939, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 3.084753605305776e-05, | |
| "eval_loss": 3.203700304031372, | |
| "eval_runtime": 2822.6209, | |
| "eval_samples_per_second": 9.672, | |
| "eval_steps_per_second": 1.209, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 6.169507210611552e-05, | |
| "grad_norm": 22.151226043701172, | |
| "learning_rate": 2e-05, | |
| "loss": 6.1671, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 9.254260815917329e-05, | |
| "grad_norm": 21.279905319213867, | |
| "learning_rate": 3e-05, | |
| "loss": 6.1115, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00012339014421223105, | |
| "grad_norm": 17.94328498840332, | |
| "learning_rate": 4e-05, | |
| "loss": 6.0745, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.0001542376802652888, | |
| "grad_norm": 18.158964157104492, | |
| "learning_rate": 5e-05, | |
| "loss": 6.3285, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0001542376802652888, | |
| "eval_loss": 2.9307665824890137, | |
| "eval_runtime": 2828.5984, | |
| "eval_samples_per_second": 9.651, | |
| "eval_steps_per_second": 1.207, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.00018508521631834657, | |
| "grad_norm": 19.19334602355957, | |
| "learning_rate": 6e-05, | |
| "loss": 5.8538, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.00021593275237140433, | |
| "grad_norm": 12.77711296081543, | |
| "learning_rate": 7e-05, | |
| "loss": 5.4956, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0002467802884244621, | |
| "grad_norm": 13.457540512084961, | |
| "learning_rate": 8e-05, | |
| "loss": 5.5774, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.00027762782447751986, | |
| "grad_norm": 12.25865650177002, | |
| "learning_rate": 9e-05, | |
| "loss": 5.3873, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0003084753605305776, | |
| "grad_norm": 10.984325408935547, | |
| "learning_rate": 0.0001, | |
| "loss": 5.3935, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0003084753605305776, | |
| "eval_loss": 2.6323626041412354, | |
| "eval_runtime": 2827.8251, | |
| "eval_samples_per_second": 9.654, | |
| "eval_steps_per_second": 1.207, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.0003393228965836354, | |
| "grad_norm": 9.817325592041016, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 4.9745, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.00037017043263669314, | |
| "grad_norm": 10.273208618164062, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 5.4002, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.0004010179686897509, | |
| "grad_norm": 9.312756538391113, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 4.9363, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.00043186550474280867, | |
| "grad_norm": 9.961502075195312, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 5.2669, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.0004627130407958664, | |
| "grad_norm": 9.298792839050293, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 5.0542, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0004627130407958664, | |
| "eval_loss": 2.485819101333618, | |
| "eval_runtime": 2827.2455, | |
| "eval_samples_per_second": 9.656, | |
| "eval_steps_per_second": 1.207, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0004935605768489242, | |
| "grad_norm": 8.941043853759766, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 4.8081, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.000524408112901982, | |
| "grad_norm": 10.766007423400879, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 4.9779, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.0005552556489550397, | |
| "grad_norm": 9.359299659729004, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 4.7944, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0005861031850080975, | |
| "grad_norm": 10.347040176391602, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 5.3346, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.0006169507210611552, | |
| "grad_norm": 9.485566139221191, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 4.9044, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.0006169507210611552, | |
| "eval_loss": 2.3902945518493652, | |
| "eval_runtime": 2828.2904, | |
| "eval_samples_per_second": 9.652, | |
| "eval_steps_per_second": 1.207, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.000647798257114213, | |
| "grad_norm": 8.43570327758789, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 4.4549, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0006786457931672708, | |
| "grad_norm": 9.576275825500488, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 4.611, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.0007094933292203286, | |
| "grad_norm": 7.116196632385254, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 4.4602, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.0007403408652733863, | |
| "grad_norm": 8.869831085205078, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 4.9386, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0007711884013264441, | |
| "grad_norm": 8.047261238098145, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 4.0805, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0007711884013264441, | |
| "eval_loss": 2.3176326751708984, | |
| "eval_runtime": 2827.886, | |
| "eval_samples_per_second": 9.654, | |
| "eval_steps_per_second": 1.207, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.0008020359373795018, | |
| "grad_norm": 7.800178527832031, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 4.8528, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.0008328834734325596, | |
| "grad_norm": 7.349645137786865, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 4.1619, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0008637310094856173, | |
| "grad_norm": 7.673518657684326, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 4.7956, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.0008945785455386751, | |
| "grad_norm": 10.641539573669434, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 4.3561, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.0009254260815917329, | |
| "grad_norm": 9.313581466674805, | |
| "learning_rate": 5e-05, | |
| "loss": 4.6394, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0009254260815917329, | |
| "eval_loss": 2.2952756881713867, | |
| "eval_runtime": 2827.877, | |
| "eval_samples_per_second": 9.654, | |
| "eval_steps_per_second": 1.207, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0009562736176447907, | |
| "grad_norm": 9.103066444396973, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 4.3205, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.0009871211536978484, | |
| "grad_norm": 8.474302291870117, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 4.5479, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.001017968689750906, | |
| "grad_norm": 8.064745903015137, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 4.4902, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.001048816225803964, | |
| "grad_norm": 10.089030265808105, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 4.7719, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.0010796637618570217, | |
| "grad_norm": 8.952753067016602, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 4.4059, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0010796637618570217, | |
| "eval_loss": 2.26672101020813, | |
| "eval_runtime": 2827.9684, | |
| "eval_samples_per_second": 9.653, | |
| "eval_steps_per_second": 1.207, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.0011105112979100794, | |
| "grad_norm": 8.924944877624512, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 4.4099, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0011413588339631371, | |
| "grad_norm": 7.8870849609375, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 3.8415, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.001172206370016195, | |
| "grad_norm": 7.664583683013916, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 4.1158, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.0012030539060692528, | |
| "grad_norm": 7.277047157287598, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 4.5537, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0012339014421223105, | |
| "grad_norm": 9.538818359375, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 4.6796, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0012339014421223105, | |
| "eval_loss": 2.2513623237609863, | |
| "eval_runtime": 2828.1493, | |
| "eval_samples_per_second": 9.653, | |
| "eval_steps_per_second": 1.207, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.0012647489781753682, | |
| "grad_norm": 8.616003036499023, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 4.4239, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.001295596514228426, | |
| "grad_norm": 7.7439398765563965, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 4.3809, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.0013264440502814838, | |
| "grad_norm": 10.247624397277832, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 5.0836, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.0013572915863345415, | |
| "grad_norm": 7.0781731605529785, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 4.4539, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.0013881391223875992, | |
| "grad_norm": 7.964367866516113, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 4.6594, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0013881391223875992, | |
| "eval_loss": 2.2453179359436035, | |
| "eval_runtime": 2827.9873, | |
| "eval_samples_per_second": 9.653, | |
| "eval_steps_per_second": 1.207, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0014189866584406572, | |
| "grad_norm": 9.104546546936035, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 4.7814, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.0014498341944937149, | |
| "grad_norm": 7.842891216278076, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 4.5211, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.0014806817305467726, | |
| "grad_norm": 7.90324068069458, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 4.689, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.0015115292665998303, | |
| "grad_norm": 7.208617687225342, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 4.5334, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.0015423768026528882, | |
| "grad_norm": 7.687088489532471, | |
| "learning_rate": 0.0, | |
| "loss": 4.5344, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.0015423768026528882, | |
| "eval_loss": 2.244182825088501, | |
| "eval_runtime": 2829.169, | |
| "eval_samples_per_second": 9.649, | |
| "eval_steps_per_second": 1.206, | |
| "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": 3.51565151993856e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |