| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.1402364451691804, | |
| "eval_steps": 500, | |
| "global_step": 31500, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.8877573609352112, | |
| "learning_rate": 9.98641119717353e-06, | |
| "loss": 2.7747, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.01, | |
| "grad_norm": 0.891560435295105, | |
| "learning_rate": 9.97282239434706e-06, | |
| "loss": 2.3805, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.02, | |
| "grad_norm": 1.140250563621521, | |
| "learning_rate": 9.959233591520588e-06, | |
| "loss": 2.2018, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 1.5957276821136475, | |
| "learning_rate": 9.945644788694116e-06, | |
| "loss": 2.1426, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "grad_norm": 1.0516589879989624, | |
| "learning_rate": 9.932055985867646e-06, | |
| "loss": 2.0379, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.03, | |
| "eval_codebleu": 0.02236215080832533, | |
| "eval_dataflow_match_score": 0.020328127538115238, | |
| "eval_loss": 2.005807876586914, | |
| "eval_ngram_match_score": 7.828601095920155e-05, | |
| "eval_runtime": 211.5866, | |
| "eval_samples_per_second": 12.019, | |
| "eval_steps_per_second": 1.503, | |
| "eval_syntax_match_score": 0.06800422172156935, | |
| "eval_weighted_ngram_match_score": 0.0010379679626575443, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.04, | |
| "grad_norm": 1.4408249855041504, | |
| "learning_rate": 9.918467183041175e-06, | |
| "loss": 2.1005, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 1.042541742324829, | |
| "learning_rate": 9.904878380214705e-06, | |
| "loss": 2.0475, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 0.05, | |
| "grad_norm": 0.9206609129905701, | |
| "learning_rate": 9.891289577388233e-06, | |
| "loss": 2.0145, | |
| "step": 800 | |
| }, | |
| { | |
| "epoch": 0.06, | |
| "grad_norm": 1.2876330614089966, | |
| "learning_rate": 9.87770077456176e-06, | |
| "loss": 1.9729, | |
| "step": 900 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 1.5496817827224731, | |
| "learning_rate": 9.86411197173529e-06, | |
| "loss": 1.9755, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "eval_codebleu": 0.026625421991617458, | |
| "eval_dataflow_match_score": 0.017334601907502377, | |
| "eval_loss": 1.8824141025543213, | |
| "eval_ngram_match_score": 0.0001360979918249884, | |
| "eval_runtime": 210.1093, | |
| "eval_samples_per_second": 12.103, | |
| "eval_steps_per_second": 1.513, | |
| "eval_syntax_match_score": 0.085743571008139, | |
| "eval_weighted_ngram_match_score": 0.0032874170590034683, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.07, | |
| "grad_norm": 1.479344129562378, | |
| "learning_rate": 9.85052316890882e-06, | |
| "loss": 1.9526, | |
| "step": 1100 | |
| }, | |
| { | |
| "epoch": 0.08, | |
| "grad_norm": 1.1165575981140137, | |
| "learning_rate": 9.83693436608235e-06, | |
| "loss": 1.9212, | |
| "step": 1200 | |
| }, | |
| { | |
| "epoch": 0.09, | |
| "grad_norm": 1.871772289276123, | |
| "learning_rate": 9.823345563255877e-06, | |
| "loss": 1.8674, | |
| "step": 1300 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 1.1310107707977295, | |
| "learning_rate": 9.809756760429407e-06, | |
| "loss": 1.9083, | |
| "step": 1400 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "grad_norm": 1.2779932022094727, | |
| "learning_rate": 9.796167957602935e-06, | |
| "loss": 1.8672, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.1, | |
| "eval_codebleu": 0.03411263800674984, | |
| "eval_dataflow_match_score": 0.035922307567354324, | |
| "eval_loss": 1.8209562301635742, | |
| "eval_ngram_match_score": 0.00019790870128407514, | |
| "eval_runtime": 209.7736, | |
| "eval_samples_per_second": 12.123, | |
| "eval_steps_per_second": 1.516, | |
| "eval_syntax_match_score": 0.09630802330065559, | |
| "eval_weighted_ngram_match_score": 0.004022312457705368, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.11, | |
| "grad_norm": 1.2445776462554932, | |
| "learning_rate": 9.782579154776465e-06, | |
| "loss": 1.8163, | |
| "step": 1600 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 1.673412561416626, | |
| "learning_rate": 9.768990351949994e-06, | |
| "loss": 1.8664, | |
| "step": 1700 | |
| }, | |
| { | |
| "epoch": 0.12, | |
| "grad_norm": 1.3831863403320312, | |
| "learning_rate": 9.755401549123524e-06, | |
| "loss": 1.8336, | |
| "step": 1800 | |
| }, | |
| { | |
| "epoch": 0.13, | |
| "grad_norm": 1.5055395364761353, | |
| "learning_rate": 9.741812746297052e-06, | |
| "loss": 1.879, | |
| "step": 1900 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 1.3736721277236938, | |
| "learning_rate": 9.72822394347058e-06, | |
| "loss": 1.8522, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "eval_codebleu": 0.04137599574546961, | |
| "eval_dataflow_match_score": 0.06186619636599912, | |
| "eval_loss": 1.7707544565200806, | |
| "eval_ngram_match_score": 0.00026554637185762897, | |
| "eval_runtime": 210.486, | |
| "eval_samples_per_second": 12.082, | |
| "eval_steps_per_second": 1.511, | |
| "eval_syntax_match_score": 0.09913941829547991, | |
| "eval_weighted_ngram_match_score": 0.00423282194854177, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.14, | |
| "grad_norm": 1.5695489645004272, | |
| "learning_rate": 9.71463514064411e-06, | |
| "loss": 1.7906, | |
| "step": 2100 | |
| }, | |
| { | |
| "epoch": 0.15, | |
| "grad_norm": 1.4151173830032349, | |
| "learning_rate": 9.701046337817639e-06, | |
| "loss": 1.8182, | |
| "step": 2200 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 1.7115099430084229, | |
| "learning_rate": 9.687457534991169e-06, | |
| "loss": 1.8232, | |
| "step": 2300 | |
| }, | |
| { | |
| "epoch": 0.16, | |
| "grad_norm": 1.1658021211624146, | |
| "learning_rate": 9.673868732164697e-06, | |
| "loss": 1.7381, | |
| "step": 2400 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "grad_norm": 1.5483379364013672, | |
| "learning_rate": 9.660279929338226e-06, | |
| "loss": 1.7707, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.17, | |
| "eval_codebleu": 0.041373696094825684, | |
| "eval_dataflow_match_score": 0.06151811199034646, | |
| "eval_loss": 1.7324126958847046, | |
| "eval_ngram_match_score": 0.00031767245966084565, | |
| "eval_runtime": 210.1473, | |
| "eval_samples_per_second": 12.101, | |
| "eval_steps_per_second": 1.513, | |
| "eval_syntax_match_score": 0.09862185146846902, | |
| "eval_weighted_ngram_match_score": 0.0050371484608264, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 1.138310194015503, | |
| "learning_rate": 9.646691126511754e-06, | |
| "loss": 1.702, | |
| "step": 2600 | |
| }, | |
| { | |
| "epoch": 0.18, | |
| "grad_norm": 1.5409811735153198, | |
| "learning_rate": 9.633102323685284e-06, | |
| "loss": 1.7403, | |
| "step": 2700 | |
| }, | |
| { | |
| "epoch": 0.19, | |
| "grad_norm": 1.2863227128982544, | |
| "learning_rate": 9.619513520858813e-06, | |
| "loss": 1.7514, | |
| "step": 2800 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 1.523219347000122, | |
| "learning_rate": 9.605924718032343e-06, | |
| "loss": 1.7735, | |
| "step": 2900 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "grad_norm": 1.6315879821777344, | |
| "learning_rate": 9.592335915205871e-06, | |
| "loss": 1.7289, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.2, | |
| "eval_codebleu": 0.03736357788677419, | |
| "eval_dataflow_match_score": 0.04515814633467152, | |
| "eval_loss": 1.6959866285324097, | |
| "eval_ngram_match_score": 0.00026766267346779954, | |
| "eval_runtime": 209.9032, | |
| "eval_samples_per_second": 12.115, | |
| "eval_steps_per_second": 1.515, | |
| "eval_syntax_match_score": 0.09879437374413931, | |
| "eval_weighted_ngram_match_score": 0.0052341287948181195, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.21, | |
| "grad_norm": 1.5990880727767944, | |
| "learning_rate": 9.5787471123794e-06, | |
| "loss": 1.7275, | |
| "step": 3100 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 1.787031650543213, | |
| "learning_rate": 9.565158309552928e-06, | |
| "loss": 1.7112, | |
| "step": 3200 | |
| }, | |
| { | |
| "epoch": 0.22, | |
| "grad_norm": 1.6794238090515137, | |
| "learning_rate": 9.551569506726458e-06, | |
| "loss": 1.683, | |
| "step": 3300 | |
| }, | |
| { | |
| "epoch": 0.23, | |
| "grad_norm": 2.1808717250823975, | |
| "learning_rate": 9.537980703899988e-06, | |
| "loss": 1.8087, | |
| "step": 3400 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 1.737121343612671, | |
| "learning_rate": 9.524391901073516e-06, | |
| "loss": 1.7739, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "eval_codebleu": 0.03697189900391692, | |
| "eval_dataflow_match_score": 0.03780196319587868, | |
| "eval_loss": 1.668063998222351, | |
| "eval_ngram_match_score": 0.0003274646611745997, | |
| "eval_runtime": 208.872, | |
| "eval_samples_per_second": 12.175, | |
| "eval_steps_per_second": 1.522, | |
| "eval_syntax_match_score": 0.1035438105096511, | |
| "eval_weighted_ngram_match_score": 0.006214357648963285, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.24, | |
| "grad_norm": 1.6818939447402954, | |
| "learning_rate": 9.510803098247045e-06, | |
| "loss": 1.7366, | |
| "step": 3600 | |
| }, | |
| { | |
| "epoch": 0.25, | |
| "grad_norm": 1.4390071630477905, | |
| "learning_rate": 9.497214295420573e-06, | |
| "loss": 1.6693, | |
| "step": 3700 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 1.443108081817627, | |
| "learning_rate": 9.483625492594103e-06, | |
| "loss": 1.6856, | |
| "step": 3800 | |
| }, | |
| { | |
| "epoch": 0.26, | |
| "grad_norm": 1.6540530920028687, | |
| "learning_rate": 9.470036689767632e-06, | |
| "loss": 1.6965, | |
| "step": 3900 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "grad_norm": 1.178989291191101, | |
| "learning_rate": 9.456447886941162e-06, | |
| "loss": 1.6797, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.27, | |
| "eval_codebleu": 0.033575238882649495, | |
| "eval_dataflow_match_score": 0.025967094423688303, | |
| "eval_loss": 1.6396487951278687, | |
| "eval_ngram_match_score": 0.00023558729997467283, | |
| "eval_runtime": 209.0765, | |
| "eval_samples_per_second": 12.163, | |
| "eval_steps_per_second": 1.521, | |
| "eval_syntax_match_score": 0.1018591812295764, | |
| "eval_weighted_ngram_match_score": 0.006239092577358598, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.28, | |
| "grad_norm": 1.888967752456665, | |
| "learning_rate": 9.44285908411469e-06, | |
| "loss": 1.6815, | |
| "step": 4100 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 1.4612641334533691, | |
| "learning_rate": 9.42927028128822e-06, | |
| "loss": 1.6089, | |
| "step": 4200 | |
| }, | |
| { | |
| "epoch": 0.29, | |
| "grad_norm": 1.5609626770019531, | |
| "learning_rate": 9.415681478461748e-06, | |
| "loss": 1.6824, | |
| "step": 4300 | |
| }, | |
| { | |
| "epoch": 0.3, | |
| "grad_norm": 1.38056218624115, | |
| "learning_rate": 9.402092675635277e-06, | |
| "loss": 1.679, | |
| "step": 4400 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 1.928477168083191, | |
| "learning_rate": 9.388503872808807e-06, | |
| "loss": 1.6737, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "eval_codebleu": 0.03436485059379589, | |
| "eval_dataflow_match_score": 0.025433365047687558, | |
| "eval_loss": 1.619603157043457, | |
| "eval_ngram_match_score": 0.00025042578946432695, | |
| "eval_runtime": 210.4293, | |
| "eval_samples_per_second": 12.085, | |
| "eval_steps_per_second": 1.511, | |
| "eval_syntax_match_score": 0.10487324686922811, | |
| "eval_weighted_ngram_match_score": 0.006902364668803559, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "grad_norm": 1.467225432395935, | |
| "learning_rate": 9.374915069982335e-06, | |
| "loss": 1.642, | |
| "step": 4600 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "grad_norm": 1.4231284856796265, | |
| "learning_rate": 9.361326267155864e-06, | |
| "loss": 1.6238, | |
| "step": 4700 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 1.61458158493042, | |
| "learning_rate": 9.347737464329394e-06, | |
| "loss": 1.6226, | |
| "step": 4800 | |
| }, | |
| { | |
| "epoch": 0.33, | |
| "grad_norm": 1.8465476036071777, | |
| "learning_rate": 9.334148661502922e-06, | |
| "loss": 1.6427, | |
| "step": 4900 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "grad_norm": 2.1492788791656494, | |
| "learning_rate": 9.320559858676452e-06, | |
| "loss": 1.6783, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.34, | |
| "eval_codebleu": 0.03197730480981573, | |
| "eval_dataflow_match_score": 0.018842967535330563, | |
| "eval_loss": 1.596398115158081, | |
| "eval_ngram_match_score": 0.00021050334214284166, | |
| "eval_runtime": 209.9696, | |
| "eval_samples_per_second": 12.111, | |
| "eval_steps_per_second": 1.515, | |
| "eval_syntax_match_score": 0.10182873612210518, | |
| "eval_weighted_ngram_match_score": 0.0070270122396843505, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 1.5661336183547974, | |
| "learning_rate": 9.306971055849981e-06, | |
| "loss": 1.6227, | |
| "step": 5100 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "grad_norm": 1.5441151857376099, | |
| "learning_rate": 9.293382253023509e-06, | |
| "loss": 1.6525, | |
| "step": 5200 | |
| }, | |
| { | |
| "epoch": 0.36, | |
| "grad_norm": 1.507628083229065, | |
| "learning_rate": 9.279793450197039e-06, | |
| "loss": 1.649, | |
| "step": 5300 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 1.5360280275344849, | |
| "learning_rate": 9.266204647370567e-06, | |
| "loss": 1.6738, | |
| "step": 5400 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "grad_norm": 1.6375775337219238, | |
| "learning_rate": 9.252615844544096e-06, | |
| "loss": 1.5854, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.37, | |
| "eval_loss": 1.5711854696273804, | |
| "eval_runtime": 70.8487, | |
| "eval_samples_per_second": 35.893, | |
| "eval_steps_per_second": 4.488, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.38, | |
| "grad_norm": 1.5210243463516235, | |
| "learning_rate": 9.239027041717626e-06, | |
| "loss": 1.621, | |
| "step": 5600 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 1.2972959280014038, | |
| "learning_rate": 9.225438238891154e-06, | |
| "loss": 1.6124, | |
| "step": 5700 | |
| }, | |
| { | |
| "epoch": 0.39, | |
| "grad_norm": 1.902424693107605, | |
| "learning_rate": 9.211849436064683e-06, | |
| "loss": 1.6012, | |
| "step": 5800 | |
| }, | |
| { | |
| "epoch": 0.4, | |
| "grad_norm": 1.865466594696045, | |
| "learning_rate": 9.198260633238213e-06, | |
| "loss": 1.5745, | |
| "step": 5900 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 1.6546335220336914, | |
| "learning_rate": 9.184671830411741e-06, | |
| "loss": 1.5874, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "eval_loss": 1.553696870803833, | |
| "eval_runtime": 70.799, | |
| "eval_samples_per_second": 35.919, | |
| "eval_steps_per_second": 4.492, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.41, | |
| "grad_norm": 1.7478700876235962, | |
| "learning_rate": 9.17108302758527e-06, | |
| "loss": 1.5666, | |
| "step": 6100 | |
| }, | |
| { | |
| "epoch": 0.42, | |
| "grad_norm": 1.5560758113861084, | |
| "learning_rate": 9.1574942247588e-06, | |
| "loss": 1.5382, | |
| "step": 6200 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 1.4764758348464966, | |
| "learning_rate": 9.143905421932328e-06, | |
| "loss": 1.5408, | |
| "step": 6300 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "grad_norm": 1.6634376049041748, | |
| "learning_rate": 9.130316619105858e-06, | |
| "loss": 1.5688, | |
| "step": 6400 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "grad_norm": 1.673550009727478, | |
| "learning_rate": 9.116727816279387e-06, | |
| "loss": 1.5445, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.44, | |
| "eval_loss": 1.5321872234344482, | |
| "eval_runtime": 70.8079, | |
| "eval_samples_per_second": 35.914, | |
| "eval_steps_per_second": 4.491, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.45, | |
| "grad_norm": 1.572766900062561, | |
| "learning_rate": 9.103139013452915e-06, | |
| "loss": 1.5596, | |
| "step": 6600 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 1.4650744199752808, | |
| "learning_rate": 9.089550210626445e-06, | |
| "loss": 1.5327, | |
| "step": 6700 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "grad_norm": 1.8133198022842407, | |
| "learning_rate": 9.075961407799973e-06, | |
| "loss": 1.5789, | |
| "step": 6800 | |
| }, | |
| { | |
| "epoch": 0.47, | |
| "grad_norm": 1.6164945363998413, | |
| "learning_rate": 9.062372604973503e-06, | |
| "loss": 1.4918, | |
| "step": 6900 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 1.7517341375350952, | |
| "learning_rate": 9.048783802147032e-06, | |
| "loss": 1.4947, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "eval_loss": 1.5142408609390259, | |
| "eval_runtime": 70.7829, | |
| "eval_samples_per_second": 35.927, | |
| "eval_steps_per_second": 4.493, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "grad_norm": 2.3255422115325928, | |
| "learning_rate": 9.03519499932056e-06, | |
| "loss": 1.5434, | |
| "step": 7100 | |
| }, | |
| { | |
| "epoch": 0.49, | |
| "grad_norm": 2.118253469467163, | |
| "learning_rate": 9.02160619649409e-06, | |
| "loss": 1.5191, | |
| "step": 7200 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 1.5948559045791626, | |
| "learning_rate": 9.008017393667618e-06, | |
| "loss": 1.5437, | |
| "step": 7300 | |
| }, | |
| { | |
| "epoch": 0.5, | |
| "grad_norm": 1.4735013246536255, | |
| "learning_rate": 8.994428590841147e-06, | |
| "loss": 1.5217, | |
| "step": 7400 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "grad_norm": 1.5940884351730347, | |
| "learning_rate": 8.980839788014677e-06, | |
| "loss": 1.5113, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.51, | |
| "eval_loss": 1.4991660118103027, | |
| "eval_runtime": 70.7857, | |
| "eval_samples_per_second": 35.925, | |
| "eval_steps_per_second": 4.492, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 1.7738714218139648, | |
| "learning_rate": 8.967250985188207e-06, | |
| "loss": 1.5253, | |
| "step": 7600 | |
| }, | |
| { | |
| "epoch": 0.52, | |
| "grad_norm": 1.8445935249328613, | |
| "learning_rate": 8.953662182361734e-06, | |
| "loss": 1.5359, | |
| "step": 7700 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "grad_norm": 1.5108485221862793, | |
| "learning_rate": 8.940073379535264e-06, | |
| "loss": 1.5228, | |
| "step": 7800 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 1.7147557735443115, | |
| "learning_rate": 8.926484576708792e-06, | |
| "loss": 1.5228, | |
| "step": 7900 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "grad_norm": 1.5899161100387573, | |
| "learning_rate": 8.912895773882322e-06, | |
| "loss": 1.4889, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.54, | |
| "eval_loss": 1.4857795238494873, | |
| "eval_runtime": 70.8945, | |
| "eval_samples_per_second": 35.87, | |
| "eval_steps_per_second": 4.486, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.55, | |
| "grad_norm": 1.6886754035949707, | |
| "learning_rate": 8.899306971055851e-06, | |
| "loss": 1.4898, | |
| "step": 8100 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 1.5048600435256958, | |
| "learning_rate": 8.885718168229381e-06, | |
| "loss": 1.5239, | |
| "step": 8200 | |
| }, | |
| { | |
| "epoch": 0.56, | |
| "grad_norm": 1.8773216009140015, | |
| "learning_rate": 8.872129365402909e-06, | |
| "loss": 1.4889, | |
| "step": 8300 | |
| }, | |
| { | |
| "epoch": 0.57, | |
| "grad_norm": 1.7063783407211304, | |
| "learning_rate": 8.858540562576437e-06, | |
| "loss": 1.4472, | |
| "step": 8400 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 1.5942984819412231, | |
| "learning_rate": 8.844951759749966e-06, | |
| "loss": 1.4998, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "eval_loss": 1.4711333513259888, | |
| "eval_runtime": 70.9176, | |
| "eval_samples_per_second": 35.858, | |
| "eval_steps_per_second": 4.484, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.58, | |
| "grad_norm": 1.818272352218628, | |
| "learning_rate": 8.831362956923496e-06, | |
| "loss": 1.5504, | |
| "step": 8600 | |
| }, | |
| { | |
| "epoch": 0.59, | |
| "grad_norm": 1.8230923414230347, | |
| "learning_rate": 8.817774154097026e-06, | |
| "loss": 1.5182, | |
| "step": 8700 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 1.8054814338684082, | |
| "learning_rate": 8.804185351270554e-06, | |
| "loss": 1.5136, | |
| "step": 8800 | |
| }, | |
| { | |
| "epoch": 0.6, | |
| "grad_norm": 1.7768468856811523, | |
| "learning_rate": 8.790596548444083e-06, | |
| "loss": 1.4643, | |
| "step": 8900 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "grad_norm": 1.9298087358474731, | |
| "learning_rate": 8.777007745617611e-06, | |
| "loss": 1.4449, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.61, | |
| "eval_loss": 1.4553042650222778, | |
| "eval_runtime": 71.0741, | |
| "eval_samples_per_second": 35.78, | |
| "eval_steps_per_second": 4.474, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "grad_norm": 1.979022741317749, | |
| "learning_rate": 8.76341894279114e-06, | |
| "loss": 1.5057, | |
| "step": 9100 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 1.6144100427627563, | |
| "learning_rate": 8.74983013996467e-06, | |
| "loss": 1.4893, | |
| "step": 9200 | |
| }, | |
| { | |
| "epoch": 0.63, | |
| "grad_norm": 1.7385257482528687, | |
| "learning_rate": 8.7362413371382e-06, | |
| "loss": 1.4427, | |
| "step": 9300 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "grad_norm": 2.218280792236328, | |
| "learning_rate": 8.722652534311728e-06, | |
| "loss": 1.445, | |
| "step": 9400 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 1.760903000831604, | |
| "learning_rate": 8.709063731485256e-06, | |
| "loss": 1.4364, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "eval_loss": 1.4402815103530884, | |
| "eval_runtime": 70.7715, | |
| "eval_samples_per_second": 35.933, | |
| "eval_steps_per_second": 4.493, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.65, | |
| "grad_norm": 1.4513877630233765, | |
| "learning_rate": 8.695474928658785e-06, | |
| "loss": 1.437, | |
| "step": 9600 | |
| }, | |
| { | |
| "epoch": 0.66, | |
| "grad_norm": 1.703837513923645, | |
| "learning_rate": 8.681886125832315e-06, | |
| "loss": 1.4565, | |
| "step": 9700 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 2.192049980163574, | |
| "learning_rate": 8.668297323005845e-06, | |
| "loss": 1.4654, | |
| "step": 9800 | |
| }, | |
| { | |
| "epoch": 0.67, | |
| "grad_norm": 2.012014150619507, | |
| "learning_rate": 8.654708520179373e-06, | |
| "loss": 1.4975, | |
| "step": 9900 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "grad_norm": 2.117527484893799, | |
| "learning_rate": 8.641119717352902e-06, | |
| "loss": 1.4446, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.68, | |
| "eval_loss": 1.4303960800170898, | |
| "eval_runtime": 70.9752, | |
| "eval_samples_per_second": 35.829, | |
| "eval_steps_per_second": 4.48, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 2.3108479976654053, | |
| "learning_rate": 8.62753091452643e-06, | |
| "loss": 1.4801, | |
| "step": 10100 | |
| }, | |
| { | |
| "epoch": 0.69, | |
| "grad_norm": 1.4589275121688843, | |
| "learning_rate": 8.61394211169996e-06, | |
| "loss": 1.4579, | |
| "step": 10200 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "grad_norm": 1.7688006162643433, | |
| "learning_rate": 8.60035330887349e-06, | |
| "loss": 1.471, | |
| "step": 10300 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 1.6766855716705322, | |
| "learning_rate": 8.586764506047019e-06, | |
| "loss": 1.4532, | |
| "step": 10400 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "grad_norm": 2.0386102199554443, | |
| "learning_rate": 8.573175703220547e-06, | |
| "loss": 1.3998, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.71, | |
| "eval_loss": 1.4204617738723755, | |
| "eval_runtime": 70.7476, | |
| "eval_samples_per_second": 35.945, | |
| "eval_steps_per_second": 4.495, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 0.72, | |
| "grad_norm": 1.9797570705413818, | |
| "learning_rate": 8.559586900394075e-06, | |
| "loss": 1.3922, | |
| "step": 10600 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 1.7562373876571655, | |
| "learning_rate": 8.545998097567605e-06, | |
| "loss": 1.4378, | |
| "step": 10700 | |
| }, | |
| { | |
| "epoch": 0.73, | |
| "grad_norm": 1.6127831935882568, | |
| "learning_rate": 8.532409294741134e-06, | |
| "loss": 1.4483, | |
| "step": 10800 | |
| }, | |
| { | |
| "epoch": 0.74, | |
| "grad_norm": 1.6120541095733643, | |
| "learning_rate": 8.518820491914664e-06, | |
| "loss": 1.3961, | |
| "step": 10900 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 1.5521306991577148, | |
| "learning_rate": 8.505231689088192e-06, | |
| "loss": 1.4101, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "eval_loss": 1.4052175283432007, | |
| "eval_runtime": 70.9316, | |
| "eval_samples_per_second": 35.851, | |
| "eval_steps_per_second": 4.483, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 0.75, | |
| "grad_norm": 2.4100379943847656, | |
| "learning_rate": 8.491642886261721e-06, | |
| "loss": 1.4224, | |
| "step": 11100 | |
| }, | |
| { | |
| "epoch": 0.76, | |
| "grad_norm": 1.7542225122451782, | |
| "learning_rate": 8.47805408343525e-06, | |
| "loss": 1.4238, | |
| "step": 11200 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 2.3809213638305664, | |
| "learning_rate": 8.464465280608779e-06, | |
| "loss": 1.3968, | |
| "step": 11300 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "grad_norm": 1.490343451499939, | |
| "learning_rate": 8.450876477782309e-06, | |
| "loss": 1.4512, | |
| "step": 11400 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "grad_norm": 1.663609504699707, | |
| "learning_rate": 8.437287674955838e-06, | |
| "loss": 1.4772, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.78, | |
| "eval_loss": 1.3936774730682373, | |
| "eval_runtime": 70.914, | |
| "eval_samples_per_second": 35.86, | |
| "eval_steps_per_second": 4.484, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 1.5684208869934082, | |
| "learning_rate": 8.423698872129366e-06, | |
| "loss": 1.4276, | |
| "step": 11600 | |
| }, | |
| { | |
| "epoch": 0.79, | |
| "grad_norm": 1.6131608486175537, | |
| "learning_rate": 8.410110069302894e-06, | |
| "loss": 1.4067, | |
| "step": 11700 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "grad_norm": 2.017564058303833, | |
| "learning_rate": 8.396521266476424e-06, | |
| "loss": 1.4028, | |
| "step": 11800 | |
| }, | |
| { | |
| "epoch": 0.81, | |
| "grad_norm": 2.383514165878296, | |
| "learning_rate": 8.382932463649953e-06, | |
| "loss": 1.4028, | |
| "step": 11900 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 2.202026605606079, | |
| "learning_rate": 8.369343660823483e-06, | |
| "loss": 1.3671, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "eval_loss": 1.3839792013168335, | |
| "eval_runtime": 70.7236, | |
| "eval_samples_per_second": 35.957, | |
| "eval_steps_per_second": 4.496, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 0.82, | |
| "grad_norm": 1.499125361442566, | |
| "learning_rate": 8.355754857997011e-06, | |
| "loss": 1.4193, | |
| "step": 12100 | |
| }, | |
| { | |
| "epoch": 0.83, | |
| "grad_norm": 1.3109521865844727, | |
| "learning_rate": 8.34216605517054e-06, | |
| "loss": 1.3969, | |
| "step": 12200 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 2.689412832260132, | |
| "learning_rate": 8.328577252344068e-06, | |
| "loss": 1.4141, | |
| "step": 12300 | |
| }, | |
| { | |
| "epoch": 0.84, | |
| "grad_norm": 1.6615593433380127, | |
| "learning_rate": 8.314988449517598e-06, | |
| "loss": 1.3512, | |
| "step": 12400 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "grad_norm": 1.994040846824646, | |
| "learning_rate": 8.301399646691128e-06, | |
| "loss": 1.4268, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_loss": 1.3757482767105103, | |
| "eval_runtime": 70.9944, | |
| "eval_samples_per_second": 35.82, | |
| "eval_steps_per_second": 4.479, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 2.2422096729278564, | |
| "learning_rate": 8.287810843864657e-06, | |
| "loss": 1.3549, | |
| "step": 12600 | |
| }, | |
| { | |
| "epoch": 0.86, | |
| "grad_norm": 1.4407843351364136, | |
| "learning_rate": 8.274222041038185e-06, | |
| "loss": 1.3934, | |
| "step": 12700 | |
| }, | |
| { | |
| "epoch": 0.87, | |
| "grad_norm": 3.7289652824401855, | |
| "learning_rate": 8.260633238211713e-06, | |
| "loss": 1.3916, | |
| "step": 12800 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 1.819023847579956, | |
| "learning_rate": 8.247044435385243e-06, | |
| "loss": 1.3878, | |
| "step": 12900 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "grad_norm": 1.6075499057769775, | |
| "learning_rate": 8.233455632558772e-06, | |
| "loss": 1.3469, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "eval_loss": 1.365922212600708, | |
| "eval_runtime": 71.0057, | |
| "eval_samples_per_second": 35.814, | |
| "eval_steps_per_second": 4.479, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 0.89, | |
| "grad_norm": 1.9204126596450806, | |
| "learning_rate": 8.219866829732302e-06, | |
| "loss": 1.3456, | |
| "step": 13100 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 2.0110292434692383, | |
| "learning_rate": 8.20627802690583e-06, | |
| "loss": 1.3921, | |
| "step": 13200 | |
| }, | |
| { | |
| "epoch": 0.9, | |
| "grad_norm": 1.4502629041671753, | |
| "learning_rate": 8.192689224079358e-06, | |
| "loss": 1.383, | |
| "step": 13300 | |
| }, | |
| { | |
| "epoch": 0.91, | |
| "grad_norm": 2.5011653900146484, | |
| "learning_rate": 8.179100421252888e-06, | |
| "loss": 1.3413, | |
| "step": 13400 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 1.4338220357894897, | |
| "learning_rate": 8.165511618426417e-06, | |
| "loss": 1.3531, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "eval_loss": 1.3567384481430054, | |
| "eval_runtime": 70.8084, | |
| "eval_samples_per_second": 35.914, | |
| "eval_steps_per_second": 4.491, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "grad_norm": 1.8867733478546143, | |
| "learning_rate": 8.151922815599947e-06, | |
| "loss": 1.4115, | |
| "step": 13600 | |
| }, | |
| { | |
| "epoch": 0.93, | |
| "grad_norm": 1.897558331489563, | |
| "learning_rate": 8.138334012773476e-06, | |
| "loss": 1.3473, | |
| "step": 13700 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 2.6677191257476807, | |
| "learning_rate": 8.124745209947004e-06, | |
| "loss": 1.3982, | |
| "step": 13800 | |
| }, | |
| { | |
| "epoch": 0.94, | |
| "grad_norm": 1.6690632104873657, | |
| "learning_rate": 8.111156407120532e-06, | |
| "loss": 1.3371, | |
| "step": 13900 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "grad_norm": 1.668286919593811, | |
| "learning_rate": 8.097567604294062e-06, | |
| "loss": 1.3463, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.95, | |
| "eval_loss": 1.3470206260681152, | |
| "eval_runtime": 70.8694, | |
| "eval_samples_per_second": 35.883, | |
| "eval_steps_per_second": 4.487, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.3303313255310059, | |
| "learning_rate": 8.083978801467592e-06, | |
| "loss": 1.3283, | |
| "step": 14100 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "grad_norm": 1.8314011096954346, | |
| "learning_rate": 8.070389998641121e-06, | |
| "loss": 1.3382, | |
| "step": 14200 | |
| }, | |
| { | |
| "epoch": 0.97, | |
| "grad_norm": 1.6911287307739258, | |
| "learning_rate": 8.056801195814649e-06, | |
| "loss": 1.3249, | |
| "step": 14300 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "grad_norm": 2.0255990028381348, | |
| "learning_rate": 8.043212392988179e-06, | |
| "loss": 1.4345, | |
| "step": 14400 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 1.6872771978378296, | |
| "learning_rate": 8.029623590161707e-06, | |
| "loss": 1.3662, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "eval_loss": 1.3394687175750732, | |
| "eval_runtime": 70.8165, | |
| "eval_samples_per_second": 35.91, | |
| "eval_steps_per_second": 4.49, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 0.99, | |
| "grad_norm": 1.2456538677215576, | |
| "learning_rate": 8.016034787335236e-06, | |
| "loss": 1.3152, | |
| "step": 14600 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "grad_norm": 1.9343585968017578, | |
| "learning_rate": 8.002445984508766e-06, | |
| "loss": 1.3179, | |
| "step": 14700 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 1.6026442050933838, | |
| "learning_rate": 7.988857181682294e-06, | |
| "loss": 1.3445, | |
| "step": 14800 | |
| }, | |
| { | |
| "epoch": 1.01, | |
| "grad_norm": 1.8159044981002808, | |
| "learning_rate": 7.975268378855823e-06, | |
| "loss": 1.3259, | |
| "step": 14900 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "grad_norm": 1.6430504322052002, | |
| "learning_rate": 7.961679576029351e-06, | |
| "loss": 1.337, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.02, | |
| "eval_loss": 1.3323568105697632, | |
| "eval_runtime": 70.9018, | |
| "eval_samples_per_second": 35.867, | |
| "eval_steps_per_second": 4.485, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 2.036970853805542, | |
| "learning_rate": 7.948090773202881e-06, | |
| "loss": 1.3217, | |
| "step": 15100 | |
| }, | |
| { | |
| "epoch": 1.03, | |
| "grad_norm": 1.6756584644317627, | |
| "learning_rate": 7.93450197037641e-06, | |
| "loss": 1.3008, | |
| "step": 15200 | |
| }, | |
| { | |
| "epoch": 1.04, | |
| "grad_norm": 1.5923326015472412, | |
| "learning_rate": 7.92091316754994e-06, | |
| "loss": 1.3347, | |
| "step": 15300 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 1.809383749961853, | |
| "learning_rate": 7.907460252751734e-06, | |
| "loss": 1.3192, | |
| "step": 15400 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "grad_norm": 2.035680055618286, | |
| "learning_rate": 7.893871449925262e-06, | |
| "loss": 1.3627, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "eval_loss": 1.3219527006149292, | |
| "eval_runtime": 70.9632, | |
| "eval_samples_per_second": 35.835, | |
| "eval_steps_per_second": 4.481, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.06, | |
| "grad_norm": 1.7485100030899048, | |
| "learning_rate": 7.880282647098791e-06, | |
| "loss": 1.3015, | |
| "step": 15600 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 2.0771241188049316, | |
| "learning_rate": 7.86669384427232e-06, | |
| "loss": 1.3261, | |
| "step": 15700 | |
| }, | |
| { | |
| "epoch": 1.07, | |
| "grad_norm": 1.8625783920288086, | |
| "learning_rate": 7.853105041445849e-06, | |
| "loss": 1.3308, | |
| "step": 15800 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "grad_norm": 1.8347725868225098, | |
| "learning_rate": 7.839516238619378e-06, | |
| "loss": 1.3637, | |
| "step": 15900 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "grad_norm": 1.9449338912963867, | |
| "learning_rate": 7.825927435792908e-06, | |
| "loss": 1.2906, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "eval_loss": 1.3169829845428467, | |
| "eval_runtime": 70.9704, | |
| "eval_samples_per_second": 35.832, | |
| "eval_steps_per_second": 4.481, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.09, | |
| "grad_norm": 1.6746830940246582, | |
| "learning_rate": 7.812338632966436e-06, | |
| "loss": 1.3326, | |
| "step": 16100 | |
| }, | |
| { | |
| "epoch": 1.1, | |
| "grad_norm": 1.5581905841827393, | |
| "learning_rate": 7.798749830139966e-06, | |
| "loss": 1.2964, | |
| "step": 16200 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "grad_norm": 2.1636734008789062, | |
| "learning_rate": 7.785161027313495e-06, | |
| "loss": 1.2867, | |
| "step": 16300 | |
| }, | |
| { | |
| "epoch": 1.11, | |
| "grad_norm": 1.760335922241211, | |
| "learning_rate": 7.771572224487023e-06, | |
| "loss": 1.3017, | |
| "step": 16400 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "grad_norm": 1.9209500551223755, | |
| "learning_rate": 7.757983421660553e-06, | |
| "loss": 1.331, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "eval_loss": 1.3085339069366455, | |
| "eval_runtime": 70.9762, | |
| "eval_samples_per_second": 35.829, | |
| "eval_steps_per_second": 4.48, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "grad_norm": 1.8936728239059448, | |
| "learning_rate": 7.74439461883408e-06, | |
| "loss": 1.3494, | |
| "step": 16600 | |
| }, | |
| { | |
| "epoch": 1.13, | |
| "grad_norm": 1.6603670120239258, | |
| "learning_rate": 7.73080581600761e-06, | |
| "loss": 1.3443, | |
| "step": 16700 | |
| }, | |
| { | |
| "epoch": 1.14, | |
| "grad_norm": 1.9962695837020874, | |
| "learning_rate": 7.71721701318114e-06, | |
| "loss": 1.3016, | |
| "step": 16800 | |
| }, | |
| { | |
| "epoch": 1.15, | |
| "grad_norm": 1.7902451753616333, | |
| "learning_rate": 7.70362821035467e-06, | |
| "loss": 1.3107, | |
| "step": 16900 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 1.7962889671325684, | |
| "learning_rate": 7.690039407528197e-06, | |
| "loss": 1.3082, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "eval_loss": 1.3007583618164062, | |
| "eval_runtime": 70.8534, | |
| "eval_samples_per_second": 35.891, | |
| "eval_steps_per_second": 4.488, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.16, | |
| "grad_norm": 1.8448220491409302, | |
| "learning_rate": 7.676450604701725e-06, | |
| "loss": 1.3148, | |
| "step": 17100 | |
| }, | |
| { | |
| "epoch": 1.17, | |
| "grad_norm": 2.124708652496338, | |
| "learning_rate": 7.662861801875255e-06, | |
| "loss": 1.3348, | |
| "step": 17200 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 1.5953021049499512, | |
| "learning_rate": 7.649272999048785e-06, | |
| "loss": 1.2575, | |
| "step": 17300 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "grad_norm": 1.7431753873825073, | |
| "learning_rate": 7.635684196222314e-06, | |
| "loss": 1.3239, | |
| "step": 17400 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "grad_norm": 2.0628209114074707, | |
| "learning_rate": 7.622095393395842e-06, | |
| "loss": 1.2904, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.19, | |
| "eval_loss": 1.2971383333206177, | |
| "eval_runtime": 70.8465, | |
| "eval_samples_per_second": 35.894, | |
| "eval_steps_per_second": 4.489, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 2.154141902923584, | |
| "learning_rate": 7.608506590569371e-06, | |
| "loss": 1.2824, | |
| "step": 17600 | |
| }, | |
| { | |
| "epoch": 1.2, | |
| "grad_norm": 1.7325314283370972, | |
| "learning_rate": 7.595053675771166e-06, | |
| "loss": 1.2587, | |
| "step": 17700 | |
| }, | |
| { | |
| "epoch": 1.21, | |
| "grad_norm": 1.7533044815063477, | |
| "learning_rate": 7.581464872944694e-06, | |
| "loss": 1.2697, | |
| "step": 17800 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "grad_norm": 1.642408847808838, | |
| "learning_rate": 7.567876070118223e-06, | |
| "loss": 1.2587, | |
| "step": 17900 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "grad_norm": 2.0030946731567383, | |
| "learning_rate": 7.554287267291752e-06, | |
| "loss": 1.2825, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.22, | |
| "eval_loss": 1.2882283926010132, | |
| "eval_runtime": 70.9711, | |
| "eval_samples_per_second": 35.831, | |
| "eval_steps_per_second": 4.481, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "grad_norm": 1.828447699546814, | |
| "learning_rate": 7.540698464465281e-06, | |
| "loss": 1.2162, | |
| "step": 18100 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 1.9078677892684937, | |
| "learning_rate": 7.527109661638811e-06, | |
| "loss": 1.2618, | |
| "step": 18200 | |
| }, | |
| { | |
| "epoch": 1.24, | |
| "grad_norm": 1.7438205480575562, | |
| "learning_rate": 7.513520858812339e-06, | |
| "loss": 1.2972, | |
| "step": 18300 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 1.5308886766433716, | |
| "learning_rate": 7.499932055985868e-06, | |
| "loss": 1.2635, | |
| "step": 18400 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 1.7570804357528687, | |
| "learning_rate": 7.486343253159397e-06, | |
| "loss": 1.3104, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "eval_loss": 1.2821784019470215, | |
| "eval_runtime": 70.9616, | |
| "eval_samples_per_second": 35.836, | |
| "eval_steps_per_second": 4.481, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.26, | |
| "grad_norm": 1.820691466331482, | |
| "learning_rate": 7.472754450332927e-06, | |
| "loss": 1.2978, | |
| "step": 18600 | |
| }, | |
| { | |
| "epoch": 1.27, | |
| "grad_norm": 2.0996739864349365, | |
| "learning_rate": 7.4591656475064555e-06, | |
| "loss": 1.2925, | |
| "step": 18700 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 1.6602342128753662, | |
| "learning_rate": 7.445576844679985e-06, | |
| "loss": 1.243, | |
| "step": 18800 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "grad_norm": 2.034649133682251, | |
| "learning_rate": 7.431988041853513e-06, | |
| "loss": 1.2775, | |
| "step": 18900 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "grad_norm": 1.898582100868225, | |
| "learning_rate": 7.418399239027042e-06, | |
| "loss": 1.2786, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "eval_loss": 1.2745426893234253, | |
| "eval_runtime": 70.9759, | |
| "eval_samples_per_second": 35.829, | |
| "eval_steps_per_second": 4.48, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 1.4128785133361816, | |
| "learning_rate": 7.4048104362005715e-06, | |
| "loss": 1.2656, | |
| "step": 19100 | |
| }, | |
| { | |
| "epoch": 1.3, | |
| "grad_norm": 2.2971534729003906, | |
| "learning_rate": 7.3912216333741e-06, | |
| "loss": 1.2426, | |
| "step": 19200 | |
| }, | |
| { | |
| "epoch": 1.31, | |
| "grad_norm": 1.783996820449829, | |
| "learning_rate": 7.37763283054763e-06, | |
| "loss": 1.272, | |
| "step": 19300 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 1.8958848714828491, | |
| "learning_rate": 7.364044027721159e-06, | |
| "loss": 1.2168, | |
| "step": 19400 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "grad_norm": 2.1363043785095215, | |
| "learning_rate": 7.350455224894687e-06, | |
| "loss": 1.2734, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.32, | |
| "eval_loss": 1.2699941396713257, | |
| "eval_runtime": 71.0334, | |
| "eval_samples_per_second": 35.8, | |
| "eval_steps_per_second": 4.477, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.33, | |
| "grad_norm": 2.642695903778076, | |
| "learning_rate": 7.336866422068216e-06, | |
| "loss": 1.3135, | |
| "step": 19600 | |
| }, | |
| { | |
| "epoch": 1.34, | |
| "grad_norm": 1.8240422010421753, | |
| "learning_rate": 7.323277619241746e-06, | |
| "loss": 1.2924, | |
| "step": 19700 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 1.8623692989349365, | |
| "learning_rate": 7.309688816415275e-06, | |
| "loss": 1.2907, | |
| "step": 19800 | |
| }, | |
| { | |
| "epoch": 1.35, | |
| "grad_norm": 2.2778708934783936, | |
| "learning_rate": 7.296100013588804e-06, | |
| "loss": 1.2727, | |
| "step": 19900 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "grad_norm": 1.8957061767578125, | |
| "learning_rate": 7.282511210762332e-06, | |
| "loss": 1.2656, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.36, | |
| "eval_loss": 1.2644336223602295, | |
| "eval_runtime": 70.9373, | |
| "eval_samples_per_second": 35.849, | |
| "eval_steps_per_second": 4.483, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "grad_norm": 1.7855497598648071, | |
| "learning_rate": 7.268922407935861e-06, | |
| "loss": 1.2158, | |
| "step": 20100 | |
| }, | |
| { | |
| "epoch": 1.37, | |
| "grad_norm": 1.8924943208694458, | |
| "learning_rate": 7.2553336051093905e-06, | |
| "loss": 1.2753, | |
| "step": 20200 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "grad_norm": 2.0177762508392334, | |
| "learning_rate": 7.241880690311184e-06, | |
| "loss": 1.2406, | |
| "step": 20300 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "grad_norm": 2.4161458015441895, | |
| "learning_rate": 7.228291887484713e-06, | |
| "loss": 1.2453, | |
| "step": 20400 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "grad_norm": 1.7210235595703125, | |
| "learning_rate": 7.214703084658242e-06, | |
| "loss": 1.2107, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.39, | |
| "eval_loss": 1.258453607559204, | |
| "eval_runtime": 71.0477, | |
| "eval_samples_per_second": 35.793, | |
| "eval_steps_per_second": 4.476, | |
| "step": 20500 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "grad_norm": 1.5457173585891724, | |
| "learning_rate": 7.201114281831771e-06, | |
| "loss": 1.2377, | |
| "step": 20600 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "grad_norm": 2.403831720352173, | |
| "learning_rate": 7.187525479005301e-06, | |
| "loss": 1.238, | |
| "step": 20700 | |
| }, | |
| { | |
| "epoch": 1.41, | |
| "grad_norm": 2.01042103767395, | |
| "learning_rate": 7.173936676178829e-06, | |
| "loss": 1.2528, | |
| "step": 20800 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "grad_norm": 2.197006940841675, | |
| "learning_rate": 7.160347873352358e-06, | |
| "loss": 1.2395, | |
| "step": 20900 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "grad_norm": 2.503634214401245, | |
| "learning_rate": 7.146759070525887e-06, | |
| "loss": 1.2822, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "eval_loss": 1.2508896589279175, | |
| "eval_runtime": 71.0512, | |
| "eval_samples_per_second": 35.791, | |
| "eval_steps_per_second": 4.476, | |
| "step": 21000 | |
| }, | |
| { | |
| "epoch": 1.43, | |
| "grad_norm": 1.4275486469268799, | |
| "learning_rate": 7.133170267699417e-06, | |
| "loss": 1.2337, | |
| "step": 21100 | |
| }, | |
| { | |
| "epoch": 1.44, | |
| "grad_norm": 2.1461949348449707, | |
| "learning_rate": 7.11971735290121e-06, | |
| "loss": 1.2576, | |
| "step": 21200 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 1.705665111541748, | |
| "learning_rate": 7.106128550074739e-06, | |
| "loss": 1.2311, | |
| "step": 21300 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "grad_norm": 2.058223247528076, | |
| "learning_rate": 7.0925397472482685e-06, | |
| "loss": 1.2619, | |
| "step": 21400 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "grad_norm": 1.4659618139266968, | |
| "learning_rate": 7.078950944421797e-06, | |
| "loss": 1.2188, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "eval_loss": 1.2478315830230713, | |
| "eval_runtime": 70.8088, | |
| "eval_samples_per_second": 35.914, | |
| "eval_steps_per_second": 4.491, | |
| "step": 21500 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "grad_norm": 1.7102398872375488, | |
| "learning_rate": 7.065362141595325e-06, | |
| "loss": 1.2553, | |
| "step": 21600 | |
| }, | |
| { | |
| "epoch": 1.47, | |
| "grad_norm": 2.445326089859009, | |
| "learning_rate": 7.051773338768855e-06, | |
| "loss": 1.2687, | |
| "step": 21700 | |
| }, | |
| { | |
| "epoch": 1.48, | |
| "grad_norm": 2.055088758468628, | |
| "learning_rate": 7.038184535942384e-06, | |
| "loss": 1.254, | |
| "step": 21800 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "grad_norm": 2.5781538486480713, | |
| "learning_rate": 7.024595733115913e-06, | |
| "loss": 1.2319, | |
| "step": 21900 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "grad_norm": 1.9507685899734497, | |
| "learning_rate": 7.011006930289442e-06, | |
| "loss": 1.2185, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.49, | |
| "eval_loss": 1.245086908340454, | |
| "eval_runtime": 71.042, | |
| "eval_samples_per_second": 35.796, | |
| "eval_steps_per_second": 4.476, | |
| "step": 22000 | |
| }, | |
| { | |
| "epoch": 1.5, | |
| "grad_norm": 2.347245216369629, | |
| "learning_rate": 6.997418127462972e-06, | |
| "loss": 1.2024, | |
| "step": 22100 | |
| }, | |
| { | |
| "epoch": 1.51, | |
| "grad_norm": 1.6178568601608276, | |
| "learning_rate": 6.9838293246364995e-06, | |
| "loss": 1.2379, | |
| "step": 22200 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 1.7551498413085938, | |
| "learning_rate": 6.970240521810029e-06, | |
| "loss": 1.2344, | |
| "step": 22300 | |
| }, | |
| { | |
| "epoch": 1.52, | |
| "grad_norm": 1.9250737428665161, | |
| "learning_rate": 6.956651718983558e-06, | |
| "loss": 1.2942, | |
| "step": 22400 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "grad_norm": 1.640271782875061, | |
| "learning_rate": 6.9430629161570876e-06, | |
| "loss": 1.2441, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 1.53, | |
| "eval_loss": 1.2352341413497925, | |
| "eval_runtime": 70.9437, | |
| "eval_samples_per_second": 35.845, | |
| "eval_steps_per_second": 4.482, | |
| "step": 22500 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "grad_norm": 2.083061456680298, | |
| "learning_rate": 6.929474113330616e-06, | |
| "loss": 1.1819, | |
| "step": 22600 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "grad_norm": 1.8274168968200684, | |
| "learning_rate": 6.915885310504146e-06, | |
| "loss": 1.2243, | |
| "step": 22700 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "grad_norm": 1.7711529731750488, | |
| "learning_rate": 6.902296507677674e-06, | |
| "loss": 1.2318, | |
| "step": 22800 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 2.0537028312683105, | |
| "learning_rate": 6.888707704851203e-06, | |
| "loss": 1.1958, | |
| "step": 22900 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "grad_norm": 1.8466728925704956, | |
| "learning_rate": 6.875118902024732e-06, | |
| "loss": 1.2564, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.56, | |
| "eval_loss": 1.2326687574386597, | |
| "eval_runtime": 70.9545, | |
| "eval_samples_per_second": 35.84, | |
| "eval_steps_per_second": 4.482, | |
| "step": 23000 | |
| }, | |
| { | |
| "epoch": 1.57, | |
| "grad_norm": 1.6301406621932983, | |
| "learning_rate": 6.861530099198261e-06, | |
| "loss": 1.1904, | |
| "step": 23100 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 2.1303887367248535, | |
| "learning_rate": 6.847941296371791e-06, | |
| "loss": 1.2, | |
| "step": 23200 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "grad_norm": 2.042210340499878, | |
| "learning_rate": 6.834352493545319e-06, | |
| "loss": 1.2432, | |
| "step": 23300 | |
| }, | |
| { | |
| "epoch": 1.59, | |
| "grad_norm": 1.8403574228286743, | |
| "learning_rate": 6.820763690718848e-06, | |
| "loss": 1.2195, | |
| "step": 23400 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 1.8628817796707153, | |
| "learning_rate": 6.807174887892377e-06, | |
| "loss": 1.2032, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "eval_loss": 1.2271267175674438, | |
| "eval_runtime": 70.9472, | |
| "eval_samples_per_second": 35.844, | |
| "eval_steps_per_second": 4.482, | |
| "step": 23500 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "grad_norm": 2.2309892177581787, | |
| "learning_rate": 6.793586085065907e-06, | |
| "loss": 1.1931, | |
| "step": 23600 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "grad_norm": 1.4337612390518188, | |
| "learning_rate": 6.7799972822394354e-06, | |
| "loss": 1.2401, | |
| "step": 23700 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "grad_norm": 1.7968145608901978, | |
| "learning_rate": 6.766408479412965e-06, | |
| "loss": 1.233, | |
| "step": 23800 | |
| }, | |
| { | |
| "epoch": 1.62, | |
| "grad_norm": 1.7918980121612549, | |
| "learning_rate": 6.7529555646147584e-06, | |
| "loss": 1.1874, | |
| "step": 23900 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "grad_norm": 1.9370090961456299, | |
| "learning_rate": 6.739366761788287e-06, | |
| "loss": 1.2031, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.63, | |
| "eval_loss": 1.2228479385375977, | |
| "eval_runtime": 70.9391, | |
| "eval_samples_per_second": 35.848, | |
| "eval_steps_per_second": 4.483, | |
| "step": 24000 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 2.238128900527954, | |
| "learning_rate": 6.725777958961815e-06, | |
| "loss": 1.2207, | |
| "step": 24100 | |
| }, | |
| { | |
| "epoch": 1.64, | |
| "grad_norm": 1.9183790683746338, | |
| "learning_rate": 6.712189156135345e-06, | |
| "loss": 1.2263, | |
| "step": 24200 | |
| }, | |
| { | |
| "epoch": 1.65, | |
| "grad_norm": 2.407428026199341, | |
| "learning_rate": 6.6986003533088736e-06, | |
| "loss": 1.2424, | |
| "step": 24300 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "grad_norm": 1.837365746498108, | |
| "learning_rate": 6.685011550482403e-06, | |
| "loss": 1.1832, | |
| "step": 24400 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "grad_norm": 1.9926724433898926, | |
| "learning_rate": 6.671422747655932e-06, | |
| "loss": 1.2088, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 1.66, | |
| "eval_loss": 1.2178888320922852, | |
| "eval_runtime": 71.0012, | |
| "eval_samples_per_second": 35.816, | |
| "eval_steps_per_second": 4.479, | |
| "step": 24500 | |
| }, | |
| { | |
| "epoch": 1.67, | |
| "grad_norm": 1.6491261720657349, | |
| "learning_rate": 6.657833944829462e-06, | |
| "loss": 1.2103, | |
| "step": 24600 | |
| }, | |
| { | |
| "epoch": 1.68, | |
| "grad_norm": 1.825826644897461, | |
| "learning_rate": 6.6442451420029895e-06, | |
| "loss": 1.1911, | |
| "step": 24700 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "grad_norm": 1.788091778755188, | |
| "learning_rate": 6.630656339176519e-06, | |
| "loss": 1.2246, | |
| "step": 24800 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "grad_norm": 1.8941233158111572, | |
| "learning_rate": 6.617067536350048e-06, | |
| "loss": 1.197, | |
| "step": 24900 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "grad_norm": 2.360272169113159, | |
| "learning_rate": 6.6034787335235775e-06, | |
| "loss": 1.1925, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "eval_loss": 1.2120610475540161, | |
| "eval_runtime": 70.9625, | |
| "eval_samples_per_second": 35.836, | |
| "eval_steps_per_second": 4.481, | |
| "step": 25000 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "grad_norm": 2.0026679039001465, | |
| "learning_rate": 6.589889930697106e-06, | |
| "loss": 1.2063, | |
| "step": 25100 | |
| }, | |
| { | |
| "epoch": 1.71, | |
| "grad_norm": 1.9979290962219238, | |
| "learning_rate": 6.576301127870636e-06, | |
| "loss": 1.1784, | |
| "step": 25200 | |
| }, | |
| { | |
| "epoch": 1.72, | |
| "grad_norm": 1.682900071144104, | |
| "learning_rate": 6.562712325044164e-06, | |
| "loss": 1.1587, | |
| "step": 25300 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 2.0586678981781006, | |
| "learning_rate": 6.549123522217693e-06, | |
| "loss": 1.2031, | |
| "step": 25400 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "grad_norm": 2.5424463748931885, | |
| "learning_rate": 6.535534719391222e-06, | |
| "loss": 1.2061, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 1.73, | |
| "eval_loss": 1.209425449371338, | |
| "eval_runtime": 70.8814, | |
| "eval_samples_per_second": 35.877, | |
| "eval_steps_per_second": 4.486, | |
| "step": 25500 | |
| }, | |
| { | |
| "epoch": 1.74, | |
| "grad_norm": 2.0070347785949707, | |
| "learning_rate": 6.521945916564751e-06, | |
| "loss": 1.2175, | |
| "step": 25600 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 1.7913732528686523, | |
| "learning_rate": 6.508357113738281e-06, | |
| "loss": 1.2005, | |
| "step": 25700 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "grad_norm": 1.552306890487671, | |
| "learning_rate": 6.494768310911809e-06, | |
| "loss": 1.2011, | |
| "step": 25800 | |
| }, | |
| { | |
| "epoch": 1.76, | |
| "grad_norm": 3.1545894145965576, | |
| "learning_rate": 6.481179508085338e-06, | |
| "loss": 1.17, | |
| "step": 25900 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 1.7653687000274658, | |
| "learning_rate": 6.467590705258867e-06, | |
| "loss": 1.1984, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "eval_loss": 1.2038514614105225, | |
| "eval_runtime": 70.8478, | |
| "eval_samples_per_second": 35.894, | |
| "eval_steps_per_second": 4.488, | |
| "step": 26000 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "grad_norm": 1.9187010526657104, | |
| "learning_rate": 6.454001902432397e-06, | |
| "loss": 1.2051, | |
| "step": 26100 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "grad_norm": 1.6188615560531616, | |
| "learning_rate": 6.440413099605925e-06, | |
| "loss": 1.1712, | |
| "step": 26200 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "grad_norm": 1.9360331296920776, | |
| "learning_rate": 6.426824296779455e-06, | |
| "loss": 1.1531, | |
| "step": 26300 | |
| }, | |
| { | |
| "epoch": 1.79, | |
| "grad_norm": 2.710357189178467, | |
| "learning_rate": 6.413235493952983e-06, | |
| "loss": 1.1707, | |
| "step": 26400 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 2.3331565856933594, | |
| "learning_rate": 6.399646691126512e-06, | |
| "loss": 1.1929, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "eval_loss": 1.2011253833770752, | |
| "eval_runtime": 70.8706, | |
| "eval_samples_per_second": 35.882, | |
| "eval_steps_per_second": 4.487, | |
| "step": 26500 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "grad_norm": 2.030912399291992, | |
| "learning_rate": 6.386057888300041e-06, | |
| "loss": 1.1986, | |
| "step": 26600 | |
| }, | |
| { | |
| "epoch": 1.81, | |
| "grad_norm": 2.1584174633026123, | |
| "learning_rate": 6.37246908547357e-06, | |
| "loss": 1.2564, | |
| "step": 26700 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "grad_norm": 2.3068361282348633, | |
| "learning_rate": 6.3588802826471e-06, | |
| "loss": 1.1933, | |
| "step": 26800 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "grad_norm": 1.5643947124481201, | |
| "learning_rate": 6.3452914798206285e-06, | |
| "loss": 1.1691, | |
| "step": 26900 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "grad_norm": 2.531083822250366, | |
| "learning_rate": 6.331702676994157e-06, | |
| "loss": 1.1387, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.83, | |
| "eval_loss": 1.1969281435012817, | |
| "eval_runtime": 70.8053, | |
| "eval_samples_per_second": 35.915, | |
| "eval_steps_per_second": 4.491, | |
| "step": 27000 | |
| }, | |
| { | |
| "epoch": 1.84, | |
| "grad_norm": 1.800430417060852, | |
| "learning_rate": 6.318113874167686e-06, | |
| "loss": 1.1646, | |
| "step": 27100 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 1.8300161361694336, | |
| "learning_rate": 6.304525071341216e-06, | |
| "loss": 1.2104, | |
| "step": 27200 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "grad_norm": 1.7128742933273315, | |
| "learning_rate": 6.2909362685147445e-06, | |
| "loss": 1.1932, | |
| "step": 27300 | |
| }, | |
| { | |
| "epoch": 1.86, | |
| "grad_norm": 1.9414857625961304, | |
| "learning_rate": 6.277347465688274e-06, | |
| "loss": 1.1818, | |
| "step": 27400 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "grad_norm": 1.9951707124710083, | |
| "learning_rate": 6.263758662861802e-06, | |
| "loss": 1.2024, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 1.87, | |
| "eval_loss": 1.1932079792022705, | |
| "eval_runtime": 70.8009, | |
| "eval_samples_per_second": 35.918, | |
| "eval_steps_per_second": 4.491, | |
| "step": 27500 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 1.9903321266174316, | |
| "learning_rate": 6.250169860035331e-06, | |
| "loss": 1.1824, | |
| "step": 27600 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "grad_norm": 2.8851983547210693, | |
| "learning_rate": 6.2365810572088605e-06, | |
| "loss": 1.1851, | |
| "step": 27700 | |
| }, | |
| { | |
| "epoch": 1.89, | |
| "grad_norm": 2.0669078826904297, | |
| "learning_rate": 6.222992254382389e-06, | |
| "loss": 1.1646, | |
| "step": 27800 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 1.9089607000350952, | |
| "learning_rate": 6.209403451555919e-06, | |
| "loss": 1.1776, | |
| "step": 27900 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "grad_norm": 2.2551538944244385, | |
| "learning_rate": 6.195814648729448e-06, | |
| "loss": 1.1909, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.9, | |
| "eval_loss": 1.1877614259719849, | |
| "eval_runtime": 70.8691, | |
| "eval_samples_per_second": 35.883, | |
| "eval_steps_per_second": 4.487, | |
| "step": 28000 | |
| }, | |
| { | |
| "epoch": 1.91, | |
| "grad_norm": 1.5612294673919678, | |
| "learning_rate": 6.182225845902976e-06, | |
| "loss": 1.1559, | |
| "step": 28100 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 2.8218579292297363, | |
| "learning_rate": 6.168637043076505e-06, | |
| "loss": 1.1652, | |
| "step": 28200 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "grad_norm": 1.9702138900756836, | |
| "learning_rate": 6.155048240250035e-06, | |
| "loss": 1.1917, | |
| "step": 28300 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "grad_norm": 2.3673105239868164, | |
| "learning_rate": 6.141459437423564e-06, | |
| "loss": 1.1336, | |
| "step": 28400 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "grad_norm": 1.8467798233032227, | |
| "learning_rate": 6.127870634597093e-06, | |
| "loss": 1.1786, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_loss": 1.1837141513824463, | |
| "eval_runtime": 70.7666, | |
| "eval_samples_per_second": 35.935, | |
| "eval_steps_per_second": 4.494, | |
| "step": 28500 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "grad_norm": 1.830837368965149, | |
| "learning_rate": 6.114281831770621e-06, | |
| "loss": 1.1908, | |
| "step": 28600 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "grad_norm": 1.7194899320602417, | |
| "learning_rate": 6.10069302894415e-06, | |
| "loss": 1.1541, | |
| "step": 28700 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 1.798368215560913, | |
| "learning_rate": 6.0871042261176796e-06, | |
| "loss": 1.1487, | |
| "step": 28800 | |
| }, | |
| { | |
| "epoch": 1.96, | |
| "grad_norm": 1.9699339866638184, | |
| "learning_rate": 6.073515423291208e-06, | |
| "loss": 1.166, | |
| "step": 28900 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "grad_norm": 2.1379947662353516, | |
| "learning_rate": 6.059926620464738e-06, | |
| "loss": 1.1724, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.97, | |
| "eval_loss": 1.181123971939087, | |
| "eval_runtime": 92.5562, | |
| "eval_samples_per_second": 27.475, | |
| "eval_steps_per_second": 3.436, | |
| "step": 29000 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 2.405134439468384, | |
| "learning_rate": 6.046337817638267e-06, | |
| "loss": 1.1651, | |
| "step": 29100 | |
| }, | |
| { | |
| "epoch": 1.98, | |
| "grad_norm": 1.8744902610778809, | |
| "learning_rate": 6.032749014811795e-06, | |
| "loss": 1.201, | |
| "step": 29200 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "grad_norm": 3.014401435852051, | |
| "learning_rate": 6.019160211985324e-06, | |
| "loss": 1.1783, | |
| "step": 29300 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 2.104191780090332, | |
| "learning_rate": 6.005571409158854e-06, | |
| "loss": 1.1479, | |
| "step": 29400 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "grad_norm": 1.9670746326446533, | |
| "learning_rate": 5.991982606332383e-06, | |
| "loss": 1.1372, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.176620602607727, | |
| "eval_runtime": 92.464, | |
| "eval_samples_per_second": 27.503, | |
| "eval_steps_per_second": 3.439, | |
| "step": 29500 | |
| }, | |
| { | |
| "epoch": 2.01, | |
| "grad_norm": 2.3720388412475586, | |
| "learning_rate": 5.978393803505912e-06, | |
| "loss": 1.1476, | |
| "step": 29600 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 2.047060251235962, | |
| "learning_rate": 5.964805000679441e-06, | |
| "loss": 1.1303, | |
| "step": 29700 | |
| }, | |
| { | |
| "epoch": 2.02, | |
| "grad_norm": 1.9792667627334595, | |
| "learning_rate": 5.951216197852969e-06, | |
| "loss": 1.1462, | |
| "step": 29800 | |
| }, | |
| { | |
| "epoch": 2.03, | |
| "grad_norm": 2.241187572479248, | |
| "learning_rate": 5.937627395026499e-06, | |
| "loss": 1.1031, | |
| "step": 29900 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "grad_norm": 2.0233969688415527, | |
| "learning_rate": 5.9240385922000274e-06, | |
| "loss": 1.1396, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.04, | |
| "eval_loss": 1.1728562116622925, | |
| "eval_runtime": 92.5707, | |
| "eval_samples_per_second": 27.471, | |
| "eval_steps_per_second": 3.435, | |
| "step": 30000 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 2.0683882236480713, | |
| "learning_rate": 5.910449789373557e-06, | |
| "loss": 1.1431, | |
| "step": 30100 | |
| }, | |
| { | |
| "epoch": 2.05, | |
| "grad_norm": 1.9208968877792358, | |
| "learning_rate": 5.896860986547086e-06, | |
| "loss": 1.1391, | |
| "step": 30200 | |
| }, | |
| { | |
| "epoch": 2.06, | |
| "grad_norm": 1.6621592044830322, | |
| "learning_rate": 5.883272183720614e-06, | |
| "loss": 1.1361, | |
| "step": 30300 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "grad_norm": 1.9728045463562012, | |
| "learning_rate": 5.869683380894143e-06, | |
| "loss": 1.1606, | |
| "step": 30400 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "grad_norm": 2.3189892768859863, | |
| "learning_rate": 5.856094578067672e-06, | |
| "loss": 1.1565, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 2.07, | |
| "eval_loss": 1.1692627668380737, | |
| "eval_runtime": 92.5367, | |
| "eval_samples_per_second": 27.481, | |
| "eval_steps_per_second": 3.436, | |
| "step": 30500 | |
| }, | |
| { | |
| "epoch": 2.08, | |
| "grad_norm": 1.936689853668213, | |
| "learning_rate": 5.842505775241202e-06, | |
| "loss": 1.1294, | |
| "step": 30600 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "grad_norm": 1.4617129564285278, | |
| "learning_rate": 5.8289169724147306e-06, | |
| "loss": 1.1591, | |
| "step": 30700 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "grad_norm": 1.5474071502685547, | |
| "learning_rate": 5.81532816958826e-06, | |
| "loss": 1.1377, | |
| "step": 30800 | |
| }, | |
| { | |
| "epoch": 2.1, | |
| "grad_norm": 1.7175779342651367, | |
| "learning_rate": 5.801739366761788e-06, | |
| "loss": 1.1532, | |
| "step": 30900 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 1.8924795389175415, | |
| "learning_rate": 5.788150563935318e-06, | |
| "loss": 1.1002, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "eval_loss": 1.1667861938476562, | |
| "eval_runtime": 92.475, | |
| "eval_samples_per_second": 27.499, | |
| "eval_steps_per_second": 3.439, | |
| "step": 31000 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "grad_norm": 2.3616528511047363, | |
| "learning_rate": 5.7745617611088465e-06, | |
| "loss": 1.1616, | |
| "step": 31100 | |
| }, | |
| { | |
| "epoch": 2.12, | |
| "grad_norm": 1.7967276573181152, | |
| "learning_rate": 5.760972958282376e-06, | |
| "loss": 1.1817, | |
| "step": 31200 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "grad_norm": 2.9053776264190674, | |
| "learning_rate": 5.747384155455905e-06, | |
| "loss": 1.1611, | |
| "step": 31300 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "grad_norm": 2.2042810916900635, | |
| "learning_rate": 5.7337953526294346e-06, | |
| "loss": 1.1394, | |
| "step": 31400 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "grad_norm": 1.8876034021377563, | |
| "learning_rate": 5.7202065498029625e-06, | |
| "loss": 1.1171, | |
| "step": 31500 | |
| }, | |
| { | |
| "epoch": 2.14, | |
| "eval_loss": 1.1626156568527222, | |
| "eval_runtime": 92.4663, | |
| "eval_samples_per_second": 27.502, | |
| "eval_steps_per_second": 3.439, | |
| "step": 31500 | |
| } | |
| ], | |
| "logging_steps": 100, | |
| "max_steps": 73590, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 500, | |
| "total_flos": 1.55848975386624e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |