| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.05412054120541206, | |
| "eval_steps": 9, | |
| "global_step": 99, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0005466721333880005, | |
| "eval_loss": 0.947159469127655, | |
| "eval_runtime": 419.8436, | |
| "eval_samples_per_second": 7.338, | |
| "eval_steps_per_second": 0.919, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0016400164001640015, | |
| "grad_norm": 0.40049853920936584, | |
| "learning_rate": 3e-05, | |
| "loss": 0.9586, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.003280032800328003, | |
| "grad_norm": 0.5728951096534729, | |
| "learning_rate": 6e-05, | |
| "loss": 0.899, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.004920049200492005, | |
| "grad_norm": 0.5787760615348816, | |
| "learning_rate": 9e-05, | |
| "loss": 0.9511, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.004920049200492005, | |
| "eval_loss": 0.8331949710845947, | |
| "eval_runtime": 422.8511, | |
| "eval_samples_per_second": 7.286, | |
| "eval_steps_per_second": 0.913, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.006560065600656006, | |
| "grad_norm": 0.3018288016319275, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.8476, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.008200082000820008, | |
| "grad_norm": 0.2912035286426544, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.6816, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.00984009840098401, | |
| "grad_norm": 0.39786526560783386, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.7029, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.00984009840098401, | |
| "eval_loss": 0.6832850575447083, | |
| "eval_runtime": 422.898, | |
| "eval_samples_per_second": 7.285, | |
| "eval_steps_per_second": 0.913, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.011480114801148012, | |
| "grad_norm": 0.23741497099399567, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.6668, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.013120131201312012, | |
| "grad_norm": 0.3272160291671753, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.6661, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.014760147601476014, | |
| "grad_norm": 0.21630150079727173, | |
| "learning_rate": 9.145187862775209e-05, | |
| "loss": 0.5646, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.014760147601476014, | |
| "eval_loss": 0.6439428329467773, | |
| "eval_runtime": 422.9121, | |
| "eval_samples_per_second": 7.285, | |
| "eval_steps_per_second": 0.913, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.016400164001640016, | |
| "grad_norm": 0.2171449512243271, | |
| "learning_rate": 8.83022221559489e-05, | |
| "loss": 0.686, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.018040180401804017, | |
| "grad_norm": 0.14807440340518951, | |
| "learning_rate": 8.473291852294987e-05, | |
| "loss": 0.5976, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.01968019680196802, | |
| "grad_norm": 0.19816875457763672, | |
| "learning_rate": 8.07830737662829e-05, | |
| "loss": 0.6343, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.01968019680196802, | |
| "eval_loss": 0.626940906047821, | |
| "eval_runtime": 422.6652, | |
| "eval_samples_per_second": 7.289, | |
| "eval_steps_per_second": 0.913, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.02132021320213202, | |
| "grad_norm": 0.17544706165790558, | |
| "learning_rate": 7.649596321166024e-05, | |
| "loss": 0.6093, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.022960229602296024, | |
| "grad_norm": 0.19559617340564728, | |
| "learning_rate": 7.191855733945387e-05, | |
| "loss": 0.6168, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.024600246002460024, | |
| "grad_norm": 0.17366176843643188, | |
| "learning_rate": 6.710100716628344e-05, | |
| "loss": 0.6461, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.024600246002460024, | |
| "eval_loss": 0.6172654628753662, | |
| "eval_runtime": 422.8704, | |
| "eval_samples_per_second": 7.286, | |
| "eval_steps_per_second": 0.913, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.026240262402624025, | |
| "grad_norm": 0.1722017079591751, | |
| "learning_rate": 6.209609477998338e-05, | |
| "loss": 0.5852, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.02788027880278803, | |
| "grad_norm": 0.18989624083042145, | |
| "learning_rate": 5.695865504800327e-05, | |
| "loss": 0.6745, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.02952029520295203, | |
| "grad_norm": 0.23755036294460297, | |
| "learning_rate": 5.174497483512506e-05, | |
| "loss": 0.6737, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.02952029520295203, | |
| "eval_loss": 0.6107719540596008, | |
| "eval_runtime": 422.7114, | |
| "eval_samples_per_second": 7.289, | |
| "eval_steps_per_second": 0.913, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.031160311603116032, | |
| "grad_norm": 0.20679989457130432, | |
| "learning_rate": 4.6512176312793736e-05, | |
| "loss": 0.5339, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.03280032800328003, | |
| "grad_norm": 0.16482852399349213, | |
| "learning_rate": 4.131759111665349e-05, | |
| "loss": 0.5692, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.03444034440344403, | |
| "grad_norm": 0.1565546691417694, | |
| "learning_rate": 3.6218132209150045e-05, | |
| "loss": 0.5936, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.03444034440344403, | |
| "eval_loss": 0.606318473815918, | |
| "eval_runtime": 422.8445, | |
| "eval_samples_per_second": 7.286, | |
| "eval_steps_per_second": 0.913, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.03608036080360803, | |
| "grad_norm": 0.1749543398618698, | |
| "learning_rate": 3.12696703292044e-05, | |
| "loss": 0.5603, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.03772037720377204, | |
| "grad_norm": 0.19363927841186523, | |
| "learning_rate": 2.6526421860705473e-05, | |
| "loss": 0.5691, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.03936039360393604, | |
| "grad_norm": 0.22506117820739746, | |
| "learning_rate": 2.2040354826462668e-05, | |
| "loss": 0.5878, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.03936039360393604, | |
| "eval_loss": 0.6035451292991638, | |
| "eval_runtime": 422.7242, | |
| "eval_samples_per_second": 7.288, | |
| "eval_steps_per_second": 0.913, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.04100041000410004, | |
| "grad_norm": 0.1970762312412262, | |
| "learning_rate": 1.7860619515673033e-05, | |
| "loss": 0.5685, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.04264042640426404, | |
| "grad_norm": 0.19608396291732788, | |
| "learning_rate": 1.4033009983067452e-05, | |
| "loss": 0.6088, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.04428044280442804, | |
| "grad_norm": 0.17808747291564941, | |
| "learning_rate": 1.0599462319663905e-05, | |
| "loss": 0.5845, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.04428044280442804, | |
| "eval_loss": 0.6018369197845459, | |
| "eval_runtime": 422.6566, | |
| "eval_samples_per_second": 7.29, | |
| "eval_steps_per_second": 0.913, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.04592045920459205, | |
| "grad_norm": 0.1903477907180786, | |
| "learning_rate": 7.597595192178702e-06, | |
| "loss": 0.6178, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.04756047560475605, | |
| "grad_norm": 0.19779753684997559, | |
| "learning_rate": 5.060297685041659e-06, | |
| "loss": 0.5633, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.04920049200492005, | |
| "grad_norm": 0.20657652616500854, | |
| "learning_rate": 3.0153689607045845e-06, | |
| "loss": 0.6083, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.04920049200492005, | |
| "eval_loss": 0.6009935736656189, | |
| "eval_runtime": 422.6386, | |
| "eval_samples_per_second": 7.29, | |
| "eval_steps_per_second": 0.913, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.05084050840508405, | |
| "grad_norm": 0.1621209979057312, | |
| "learning_rate": 1.4852136862001764e-06, | |
| "loss": 0.6047, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.05248052480524805, | |
| "grad_norm": 0.2111944705247879, | |
| "learning_rate": 4.865965629214819e-07, | |
| "loss": 0.6373, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.05412054120541206, | |
| "grad_norm": 0.21620243787765503, | |
| "learning_rate": 3.04586490452119e-08, | |
| "loss": 0.6462, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.05412054120541206, | |
| "eval_loss": 0.6008464694023132, | |
| "eval_runtime": 422.5542, | |
| "eval_samples_per_second": 7.291, | |
| "eval_steps_per_second": 0.913, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.594794408771584e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |