| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.0, | |
| "eval_steps": 500, | |
| "global_step": 20958, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.047714476572192, | |
| "grad_norm": 0.023498278111219406, | |
| "learning_rate": 0.00019523809523809525, | |
| "loss": 0.4674, | |
| "mean_token_accuracy": 0.8906204112768173, | |
| "num_tokens": 360041.0, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.095428953144384, | |
| "grad_norm": 3.823547601699829, | |
| "learning_rate": 0.00019046664758087605, | |
| "loss": 0.3194, | |
| "mean_token_accuracy": 0.9205255397558212, | |
| "num_tokens": 727772.0, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.143143429716576, | |
| "grad_norm": 0.0001839943724917248, | |
| "learning_rate": 0.00018569519992365686, | |
| "loss": 0.2674, | |
| "mean_token_accuracy": 0.9327976078987121, | |
| "num_tokens": 1097406.0, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.190857906288768, | |
| "grad_norm": 4.199777126312256, | |
| "learning_rate": 0.00018092375226643766, | |
| "loss": 0.326, | |
| "mean_token_accuracy": 0.920593403160572, | |
| "num_tokens": 1462220.0, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.23857238286096002, | |
| "grad_norm": 5.751025676727295, | |
| "learning_rate": 0.00017615230460921847, | |
| "loss": 0.2693, | |
| "mean_token_accuracy": 0.936355301618576, | |
| "num_tokens": 1828224.0, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.286286859433152, | |
| "grad_norm": 3.971045732498169, | |
| "learning_rate": 0.00017138085695199925, | |
| "loss": 0.2827, | |
| "mean_token_accuracy": 0.9261101527214051, | |
| "num_tokens": 2211342.0, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 0.33400133600534404, | |
| "grad_norm": 2.15079665184021, | |
| "learning_rate": 0.00016660940929478005, | |
| "loss": 0.2461, | |
| "mean_token_accuracy": 0.9350018633604049, | |
| "num_tokens": 2572498.0, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 0.381715812577536, | |
| "grad_norm": 0.5263189077377319, | |
| "learning_rate": 0.00016183796163756083, | |
| "loss": 0.2534, | |
| "mean_token_accuracy": 0.9374513441324234, | |
| "num_tokens": 2949312.0, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 0.42943028914972803, | |
| "grad_norm": 0.004941379185765982, | |
| "learning_rate": 0.00015706651398034164, | |
| "loss": 0.2395, | |
| "mean_token_accuracy": 0.9378743978738785, | |
| "num_tokens": 3314910.0, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 0.47714476572192005, | |
| "grad_norm": 0.007795912679284811, | |
| "learning_rate": 0.00015229506632312244, | |
| "loss": 0.2284, | |
| "mean_token_accuracy": 0.9409392136335373, | |
| "num_tokens": 3686624.0, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 0.5248592422941121, | |
| "grad_norm": 0.0033825524151325226, | |
| "learning_rate": 0.00014752361866590325, | |
| "loss": 0.2145, | |
| "mean_token_accuracy": 0.9464458491802216, | |
| "num_tokens": 4063459.0, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 0.572573718866304, | |
| "grad_norm": 3.8457958698272705, | |
| "learning_rate": 0.00014275217100868402, | |
| "loss": 0.2268, | |
| "mean_token_accuracy": 0.9410561621189117, | |
| "num_tokens": 4423993.0, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 0.620288195438496, | |
| "grad_norm": 0.020238121971488, | |
| "learning_rate": 0.00013798072335146483, | |
| "loss": 0.223, | |
| "mean_token_accuracy": 0.9393890690803528, | |
| "num_tokens": 4788203.0, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 0.6680026720106881, | |
| "grad_norm": 0.0006973391864448786, | |
| "learning_rate": 0.00013320927569424564, | |
| "loss": 0.2434, | |
| "mean_token_accuracy": 0.9404154337644577, | |
| "num_tokens": 5162496.0, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 0.71571714858288, | |
| "grad_norm": 0.0010198453674092889, | |
| "learning_rate": 0.00012843782803702644, | |
| "loss": 0.191, | |
| "mean_token_accuracy": 0.9535960764884949, | |
| "num_tokens": 5520427.0, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 0.763431625155072, | |
| "grad_norm": 0.001297333394177258, | |
| "learning_rate": 0.00012366638037980725, | |
| "loss": 0.2167, | |
| "mean_token_accuracy": 0.942195966720581, | |
| "num_tokens": 5901763.0, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 0.8111461017272641, | |
| "grad_norm": 6.195448398590088, | |
| "learning_rate": 0.00011889493272258805, | |
| "loss": 0.2305, | |
| "mean_token_accuracy": 0.9376264967918396, | |
| "num_tokens": 6272492.0, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 0.8588605782994561, | |
| "grad_norm": 0.0025545568205416203, | |
| "learning_rate": 0.00011412348506536883, | |
| "loss": 0.2303, | |
| "mean_token_accuracy": 0.9435879285335541, | |
| "num_tokens": 6657487.0, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 0.906575054871648, | |
| "grad_norm": 0.0006595577578991652, | |
| "learning_rate": 0.00010935203740814964, | |
| "loss": 0.179, | |
| "mean_token_accuracy": 0.9520990616083145, | |
| "num_tokens": 7022908.0, | |
| "step": 9500 | |
| }, | |
| { | |
| "epoch": 0.9542895314438401, | |
| "grad_norm": 3.1752443313598633, | |
| "learning_rate": 0.00010458058975093044, | |
| "loss": 0.1827, | |
| "mean_token_accuracy": 0.9488343714475632, | |
| "num_tokens": 7376243.0, | |
| "step": 10000 | |
| }, | |
| { | |
| "epoch": 1.002004008016032, | |
| "grad_norm": 0.001512572169303894, | |
| "learning_rate": 9.980914209371123e-05, | |
| "loss": 0.2101, | |
| "mean_token_accuracy": 0.9424606282711029, | |
| "num_tokens": 7749938.0, | |
| "step": 10500 | |
| }, | |
| { | |
| "epoch": 1.0497184845882241, | |
| "grad_norm": 3.992393732070923, | |
| "learning_rate": 9.503769443649203e-05, | |
| "loss": 0.1309, | |
| "mean_token_accuracy": 0.9618149808645249, | |
| "num_tokens": 8127846.0, | |
| "step": 11000 | |
| }, | |
| { | |
| "epoch": 1.097432961160416, | |
| "grad_norm": 0.00025509227998554707, | |
| "learning_rate": 9.026624677927283e-05, | |
| "loss": 0.1219, | |
| "mean_token_accuracy": 0.9635183781385421, | |
| "num_tokens": 8503956.0, | |
| "step": 11500 | |
| }, | |
| { | |
| "epoch": 1.145147437732608, | |
| "grad_norm": 0.0013997952919453382, | |
| "learning_rate": 8.549479912205364e-05, | |
| "loss": 0.1235, | |
| "mean_token_accuracy": 0.963010191321373, | |
| "num_tokens": 8862553.0, | |
| "step": 12000 | |
| }, | |
| { | |
| "epoch": 1.1928619143048, | |
| "grad_norm": 2.114091157913208, | |
| "learning_rate": 8.072335146483443e-05, | |
| "loss": 0.1271, | |
| "mean_token_accuracy": 0.9629592669010162, | |
| "num_tokens": 9233864.0, | |
| "step": 12500 | |
| }, | |
| { | |
| "epoch": 1.240576390876992, | |
| "grad_norm": 0.277444452047348, | |
| "learning_rate": 7.595190380761523e-05, | |
| "loss": 0.1187, | |
| "mean_token_accuracy": 0.9649668201208115, | |
| "num_tokens": 9596971.0, | |
| "step": 13000 | |
| }, | |
| { | |
| "epoch": 1.288290867449184, | |
| "grad_norm": 4.878781318664551, | |
| "learning_rate": 7.118045615039604e-05, | |
| "loss": 0.1226, | |
| "mean_token_accuracy": 0.9633177869319915, | |
| "num_tokens": 9966614.0, | |
| "step": 13500 | |
| }, | |
| { | |
| "epoch": 1.3360053440213762, | |
| "grad_norm": 5.269028186798096, | |
| "learning_rate": 6.640900849317683e-05, | |
| "loss": 0.1336, | |
| "mean_token_accuracy": 0.9619411797523498, | |
| "num_tokens": 10338731.0, | |
| "step": 14000 | |
| }, | |
| { | |
| "epoch": 1.3837198205935681, | |
| "grad_norm": 0.000423251127358526, | |
| "learning_rate": 6.163756083595764e-05, | |
| "loss": 0.1386, | |
| "mean_token_accuracy": 0.9611272529363633, | |
| "num_tokens": 10724381.0, | |
| "step": 14500 | |
| }, | |
| { | |
| "epoch": 1.43143429716576, | |
| "grad_norm": 0.754705548286438, | |
| "learning_rate": 5.6866113178738436e-05, | |
| "loss": 0.1182, | |
| "mean_token_accuracy": 0.9661157331466674, | |
| "num_tokens": 11094900.0, | |
| "step": 15000 | |
| }, | |
| { | |
| "epoch": 1.479148773737952, | |
| "grad_norm": 0.0017857268685474992, | |
| "learning_rate": 5.2094665521519235e-05, | |
| "loss": 0.1175, | |
| "mean_token_accuracy": 0.9679548003673554, | |
| "num_tokens": 11454717.0, | |
| "step": 15500 | |
| }, | |
| { | |
| "epoch": 1.5268632503101442, | |
| "grad_norm": 3.5908143520355225, | |
| "learning_rate": 4.732321786430003e-05, | |
| "loss": 0.11, | |
| "mean_token_accuracy": 0.9671754879951477, | |
| "num_tokens": 11832591.0, | |
| "step": 16000 | |
| }, | |
| { | |
| "epoch": 1.5745777268823362, | |
| "grad_norm": 0.010208655148744583, | |
| "learning_rate": 4.255177020708083e-05, | |
| "loss": 0.115, | |
| "mean_token_accuracy": 0.9662747744321823, | |
| "num_tokens": 12223203.0, | |
| "step": 16500 | |
| }, | |
| { | |
| "epoch": 1.6222922034545282, | |
| "grad_norm": 0.002450750907883048, | |
| "learning_rate": 3.778032254986163e-05, | |
| "loss": 0.1112, | |
| "mean_token_accuracy": 0.9668832242488861, | |
| "num_tokens": 12593144.0, | |
| "step": 17000 | |
| }, | |
| { | |
| "epoch": 1.6700066800267201, | |
| "grad_norm": 0.004513042513281107, | |
| "learning_rate": 3.300887489264243e-05, | |
| "loss": 0.1004, | |
| "mean_token_accuracy": 0.9708232057094573, | |
| "num_tokens": 12954505.0, | |
| "step": 17500 | |
| }, | |
| { | |
| "epoch": 1.7177211565989121, | |
| "grad_norm": 0.007037173956632614, | |
| "learning_rate": 2.8237427235423232e-05, | |
| "loss": 0.1114, | |
| "mean_token_accuracy": 0.9671108702421188, | |
| "num_tokens": 13307858.0, | |
| "step": 18000 | |
| }, | |
| { | |
| "epoch": 1.765435633171104, | |
| "grad_norm": 0.015288141556084156, | |
| "learning_rate": 2.3465979578204027e-05, | |
| "loss": 0.1034, | |
| "mean_token_accuracy": 0.9698293421268463, | |
| "num_tokens": 13673816.0, | |
| "step": 18500 | |
| }, | |
| { | |
| "epoch": 1.813150109743296, | |
| "grad_norm": 6.889008045196533, | |
| "learning_rate": 1.869453192098483e-05, | |
| "loss": 0.0977, | |
| "mean_token_accuracy": 0.9710470995903016, | |
| "num_tokens": 14038630.0, | |
| "step": 19000 | |
| }, | |
| { | |
| "epoch": 1.860864586315488, | |
| "grad_norm": 2.6273930072784424, | |
| "learning_rate": 1.3923084263765626e-05, | |
| "loss": 0.0936, | |
| "mean_token_accuracy": 0.9728796405792236, | |
| "num_tokens": 14401139.0, | |
| "step": 19500 | |
| }, | |
| { | |
| "epoch": 1.90857906288768, | |
| "grad_norm": 0.004859536420553923, | |
| "learning_rate": 9.151636606546427e-06, | |
| "loss": 0.0929, | |
| "mean_token_accuracy": 0.9708857105970383, | |
| "num_tokens": 14770982.0, | |
| "step": 20000 | |
| }, | |
| { | |
| "epoch": 1.9562935394598722, | |
| "grad_norm": 0.4923778176307678, | |
| "learning_rate": 4.380188949327226e-06, | |
| "loss": 0.0963, | |
| "mean_token_accuracy": 0.9713275592327117, | |
| "num_tokens": 15136348.0, | |
| "step": 20500 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 20958, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 2, | |
| "save_steps": 5000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.399363404153889e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |