| { |
| "best_global_step": 250, |
| "best_metric": 0.678814709186554, |
| "best_model_checkpoint": "./python-code-completion-ministral-8B/checkpoint-250", |
| "epoch": 1.1741176470588235, |
| "eval_steps": 50, |
| "global_step": 250, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 1.4469833582639695, |
| "epoch": 0.047058823529411764, |
| "grad_norm": 0.49369218945503235, |
| "learning_rate": 9.999578688879085e-06, |
| "loss": 1.7076, |
| "mean_token_accuracy": 0.6342830918729305, |
| "num_tokens": 25138.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 1.430517853796482, |
| "epoch": 0.09411764705882353, |
| "grad_norm": 0.51407390832901, |
| "learning_rate": 9.949107209404664e-06, |
| "loss": 1.5636, |
| "mean_token_accuracy": 0.6443205893039703, |
| "num_tokens": 47578.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 1.4371182411909103, |
| "epoch": 0.1411764705882353, |
| "grad_norm": 0.6639813184738159, |
| "learning_rate": 9.815347070757234e-06, |
| "loss": 1.4756, |
| "mean_token_accuracy": 0.6693132996559144, |
| "num_tokens": 71393.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 1.3472976341843605, |
| "epoch": 0.18823529411764706, |
| "grad_norm": 0.3675422668457031, |
| "learning_rate": 9.60054932593026e-06, |
| "loss": 1.2256, |
| "mean_token_accuracy": 0.6993286892771721, |
| "num_tokens": 96359.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 1.1503819346427917, |
| "epoch": 0.23529411764705882, |
| "grad_norm": 0.4163525104522705, |
| "learning_rate": 9.30832881198487e-06, |
| "loss": 1.0136, |
| "mean_token_accuracy": 0.7729795873165131, |
| "num_tokens": 123155.0, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.23529411764705882, |
| "eval_entropy": 1.0332347961266835, |
| "eval_loss": 0.9837512373924255, |
| "eval_mean_token_accuracy": 0.8083523658911387, |
| "eval_num_tokens": 123155.0, |
| "eval_runtime": 25.0181, |
| "eval_samples_per_second": 5.996, |
| "eval_steps_per_second": 5.996, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.8932387962937355, |
| "epoch": 0.2823529411764706, |
| "grad_norm": 0.2984226942062378, |
| "learning_rate": 8.943603315859101e-06, |
| "loss": 0.8519, |
| "mean_token_accuracy": 0.8314538225531578, |
| "num_tokens": 149237.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.7480494946241378, |
| "epoch": 0.32941176470588235, |
| "grad_norm": 0.409252405166626, |
| "learning_rate": 8.512510812798426e-06, |
| "loss": 0.7625, |
| "mean_token_accuracy": 0.853736138343811, |
| "num_tokens": 173954.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.733599005267024, |
| "epoch": 0.3764705882352941, |
| "grad_norm": 0.3177991211414337, |
| "learning_rate": 8.022306170204233e-06, |
| "loss": 0.7384, |
| "mean_token_accuracy": 0.8508381456136703, |
| "num_tokens": 202672.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 0.6522206105291843, |
| "epoch": 0.4235294117647059, |
| "grad_norm": 0.5717107057571411, |
| "learning_rate": 7.481239055273959e-06, |
| "loss": 0.6701, |
| "mean_token_accuracy": 0.8760824427008629, |
| "num_tokens": 224695.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 0.7285707335919142, |
| "epoch": 0.47058823529411764, |
| "grad_norm": 0.21871985495090485, |
| "learning_rate": 6.898415101128571e-06, |
| "loss": 0.7319, |
| "mean_token_accuracy": 0.8523965805768967, |
| "num_tokens": 251370.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.47058823529411764, |
| "eval_entropy": 0.73099374294281, |
| "eval_loss": 0.7096962332725525, |
| "eval_mean_token_accuracy": 0.8550362022717793, |
| "eval_num_tokens": 251370.0, |
| "eval_runtime": 25.1112, |
| "eval_samples_per_second": 5.973, |
| "eval_steps_per_second": 5.973, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.6658912554383278, |
| "epoch": 0.5176470588235295, |
| "grad_norm": 0.2736637592315674, |
| "learning_rate": 6.283642667866317e-06, |
| "loss": 0.6646, |
| "mean_token_accuracy": 0.8663464844226837, |
| "num_tokens": 276477.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 0.6516228564083576, |
| "epoch": 0.5647058823529412, |
| "grad_norm": 0.2847215533256531, |
| "learning_rate": 5.647267777405178e-06, |
| "loss": 0.6503, |
| "mean_token_accuracy": 0.8681117251515389, |
| "num_tokens": 302513.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 0.7009131267666817, |
| "epoch": 0.611764705882353, |
| "grad_norm": 0.22739574313163757, |
| "learning_rate": 5e-06, |
| "loss": 0.6881, |
| "mean_token_accuracy": 0.8624396547675133, |
| "num_tokens": 329679.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 0.6468490872532129, |
| "epoch": 0.6588235294117647, |
| "grad_norm": 0.3506576716899872, |
| "learning_rate": 4.352732222594823e-06, |
| "loss": 0.6462, |
| "mean_token_accuracy": 0.871060524880886, |
| "num_tokens": 353837.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 0.6587031967937946, |
| "epoch": 0.7058823529411765, |
| "grad_norm": 0.2826354205608368, |
| "learning_rate": 3.7163573321336867e-06, |
| "loss": 0.6779, |
| "mean_token_accuracy": 0.8682561233639717, |
| "num_tokens": 376786.0, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.7058823529411765, |
| "eval_entropy": 0.7037040118376414, |
| "eval_loss": 0.6883785724639893, |
| "eval_mean_token_accuracy": 0.8590577419598897, |
| "eval_num_tokens": 376786.0, |
| "eval_runtime": 25.1113, |
| "eval_samples_per_second": 5.973, |
| "eval_steps_per_second": 5.973, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.6935035854578018, |
| "epoch": 0.7529411764705882, |
| "grad_norm": 0.32049205899238586, |
| "learning_rate": 3.101584898871431e-06, |
| "loss": 0.7223, |
| "mean_token_accuracy": 0.8527676403522492, |
| "num_tokens": 404164.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 0.6355765651911497, |
| "epoch": 0.8, |
| "grad_norm": 0.27511146664619446, |
| "learning_rate": 2.518760944726042e-06, |
| "loss": 0.641, |
| "mean_token_accuracy": 0.8753799632191658, |
| "num_tokens": 427356.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 0.6331917632371187, |
| "epoch": 0.8470588235294118, |
| "grad_norm": 0.289739191532135, |
| "learning_rate": 1.977693829795769e-06, |
| "loss": 0.646, |
| "mean_token_accuracy": 0.8754393547773361, |
| "num_tokens": 450321.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 0.7034022852778434, |
| "epoch": 0.8941176470588236, |
| "grad_norm": 0.32424843311309814, |
| "learning_rate": 1.4874891872015735e-06, |
| "loss": 0.7084, |
| "mean_token_accuracy": 0.857265616953373, |
| "num_tokens": 478364.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 0.6966812558472156, |
| "epoch": 0.9411764705882353, |
| "grad_norm": 0.27991771697998047, |
| "learning_rate": 1.0563966841408995e-06, |
| "loss": 0.7108, |
| "mean_token_accuracy": 0.857306943833828, |
| "num_tokens": 505339.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.9411764705882353, |
| "eval_entropy": 0.6970382163921992, |
| "eval_loss": 0.6800833940505981, |
| "eval_mean_token_accuracy": 0.8598487377166748, |
| "eval_num_tokens": 505339.0, |
| "eval_runtime": 25.145, |
| "eval_samples_per_second": 5.965, |
| "eval_steps_per_second": 5.965, |
| "step": 200 |
| }, |
| { |
| "entropy": 0.7158303387463093, |
| "epoch": 0.9882352941176471, |
| "grad_norm": 0.41559284925460815, |
| "learning_rate": 6.916711880151305e-07, |
| "loss": 0.716, |
| "mean_token_accuracy": 0.8561157166957856, |
| "num_tokens": 533550.0, |
| "step": 210 |
| }, |
| { |
| "entropy": 0.6228710942362484, |
| "epoch": 1.0329411764705883, |
| "grad_norm": 0.5400373935699463, |
| "learning_rate": 3.9945067406974067e-07, |
| "loss": 0.6264, |
| "mean_token_accuracy": 0.8763874330018696, |
| "num_tokens": 552959.0, |
| "step": 220 |
| }, |
| { |
| "entropy": 0.6531048402190208, |
| "epoch": 1.08, |
| "grad_norm": 0.24900081753730774, |
| "learning_rate": 1.8465292924276844e-07, |
| "loss": 0.653, |
| "mean_token_accuracy": 0.8739597767591476, |
| "num_tokens": 575104.0, |
| "step": 230 |
| }, |
| { |
| "entropy": 0.6037626899778843, |
| "epoch": 1.1270588235294117, |
| "grad_norm": 0.3135730028152466, |
| "learning_rate": 5.089279059533658e-08, |
| "loss": 0.6126, |
| "mean_token_accuracy": 0.8747931569814682, |
| "num_tokens": 603807.0, |
| "step": 240 |
| }, |
| { |
| "entropy": 0.665476355701685, |
| "epoch": 1.1741176470588235, |
| "grad_norm": 0.26424089074134827, |
| "learning_rate": 4.213111209155907e-10, |
| "loss": 0.6645, |
| "mean_token_accuracy": 0.8687899306416511, |
| "num_tokens": 628367.0, |
| "step": 250 |
| }, |
| { |
| "epoch": 1.1741176470588235, |
| "eval_entropy": 0.6952528850237528, |
| "eval_loss": 0.678814709186554, |
| "eval_mean_token_accuracy": 0.8600315936406453, |
| "eval_num_tokens": 628367.0, |
| "eval_runtime": 25.1455, |
| "eval_samples_per_second": 5.965, |
| "eval_steps_per_second": 5.965, |
| "step": 250 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 250, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 2, |
| "save_steps": 100, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 2.8870417005060096e+16, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
