| { |
| "best_global_step": 200, |
| "best_metric": 0.6829276084899902, |
| "best_model_checkpoint": "./python-code-explaining-ministral-8B/checkpoint-200", |
| "epoch": 0.9411764705882353, |
| "eval_steps": 50, |
| "global_step": 200, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "entropy": 1.4653201967477798, |
| "epoch": 0.047058823529411764, |
| "grad_norm": 0.7277931571006775, |
| "learning_rate": 7.500000000000001e-06, |
| "loss": 2.0127, |
| "mean_token_accuracy": 0.5873390108346939, |
| "num_tokens": 15889.0, |
| "step": 10 |
| }, |
| { |
| "entropy": 1.4459994971752166, |
| "epoch": 0.09411764705882353, |
| "grad_norm": 0.5549399256706238, |
| "learning_rate": 9.991971098713136e-06, |
| "loss": 1.7954, |
| "mean_token_accuracy": 0.6094311743974685, |
| "num_tokens": 35503.0, |
| "step": 20 |
| }, |
| { |
| "entropy": 1.5394100606441499, |
| "epoch": 0.1411764705882353, |
| "grad_norm": 0.6845958232879639, |
| "learning_rate": 9.952707929045018e-06, |
| "loss": 1.7686, |
| "mean_token_accuracy": 0.6184274405241013, |
| "num_tokens": 51214.0, |
| "step": 30 |
| }, |
| { |
| "entropy": 1.4925685793161392, |
| "epoch": 0.18823529411764706, |
| "grad_norm": 0.5904952883720398, |
| "learning_rate": 9.88099273969759e-06, |
| "loss": 1.4545, |
| "mean_token_accuracy": 0.6649785250425339, |
| "num_tokens": 68991.0, |
| "step": 40 |
| }, |
| { |
| "entropy": 1.2551275834441185, |
| "epoch": 0.23529411764705882, |
| "grad_norm": 0.5888926982879639, |
| "learning_rate": 9.777295435775165e-06, |
| "loss": 1.2224, |
| "mean_token_accuracy": 0.7385784611105919, |
| "num_tokens": 87871.0, |
| "step": 50 |
| }, |
| { |
| "epoch": 0.23529411764705882, |
| "eval_entropy": 1.0665295275052389, |
| "eval_loss": 1.0585410594940186, |
| "eval_mean_token_accuracy": 0.7891709744930268, |
| "eval_num_tokens": 87871.0, |
| "eval_runtime": 19.7003, |
| "eval_samples_per_second": 7.614, |
| "eval_steps_per_second": 7.614, |
| "step": 50 |
| }, |
| { |
| "entropy": 0.9183753415942192, |
| "epoch": 0.2823529411764706, |
| "grad_norm": 0.4278884828090668, |
| "learning_rate": 9.64229548131959e-06, |
| "loss": 0.9356, |
| "mean_token_accuracy": 0.8226072296500206, |
| "num_tokens": 103270.0, |
| "step": 60 |
| }, |
| { |
| "entropy": 0.8008665464818477, |
| "epoch": 0.32941176470588235, |
| "grad_norm": 0.45098593831062317, |
| "learning_rate": 9.476877447204308e-06, |
| "loss": 0.8929, |
| "mean_token_accuracy": 0.8326325386762619, |
| "num_tokens": 120480.0, |
| "step": 70 |
| }, |
| { |
| "entropy": 0.7590022340416909, |
| "epoch": 0.3764705882352941, |
| "grad_norm": 0.33732372522354126, |
| "learning_rate": 9.282125215090694e-06, |
| "loss": 0.8101, |
| "mean_token_accuracy": 0.8484514832496644, |
| "num_tokens": 138964.0, |
| "step": 80 |
| }, |
| { |
| "entropy": 0.7300104968249798, |
| "epoch": 0.4235294117647059, |
| "grad_norm": 0.28080910444259644, |
| "learning_rate": 9.059314875424553e-06, |
| "loss": 0.748, |
| "mean_token_accuracy": 0.8571715533733368, |
| "num_tokens": 156022.0, |
| "step": 90 |
| }, |
| { |
| "entropy": 0.8075251087546349, |
| "epoch": 0.47058823529411764, |
| "grad_norm": 0.2512107193470001, |
| "learning_rate": 8.809906366007832e-06, |
| "loss": 0.8393, |
| "mean_token_accuracy": 0.8405888199806213, |
| "num_tokens": 175876.0, |
| "step": 100 |
| }, |
| { |
| "epoch": 0.47058823529411764, |
| "eval_entropy": 0.741940244436264, |
| "eval_loss": 0.7406004071235657, |
| "eval_mean_token_accuracy": 0.8551796130339304, |
| "eval_num_tokens": 175876.0, |
| "eval_runtime": 19.7066, |
| "eval_samples_per_second": 7.612, |
| "eval_steps_per_second": 7.612, |
| "step": 100 |
| }, |
| { |
| "entropy": 0.7609439708292485, |
| "epoch": 0.5176470588235295, |
| "grad_norm": 0.2839739918708801, |
| "learning_rate": 8.535533905932739e-06, |
| "loss": 0.7876, |
| "mean_token_accuracy": 0.8530501529574395, |
| "num_tokens": 193834.0, |
| "step": 110 |
| }, |
| { |
| "entropy": 0.7604947410523891, |
| "epoch": 0.5647058823529412, |
| "grad_norm": 0.2849130928516388, |
| "learning_rate": 8.237995287558801e-06, |
| "loss": 0.7681, |
| "mean_token_accuracy": 0.8534491807222366, |
| "num_tokens": 212048.0, |
| "step": 120 |
| }, |
| { |
| "entropy": 0.7774970278143882, |
| "epoch": 0.611764705882353, |
| "grad_norm": 0.2741145193576813, |
| "learning_rate": 7.919240096695877e-06, |
| "loss": 0.7804, |
| "mean_token_accuracy": 0.846380078792572, |
| "num_tokens": 231794.0, |
| "step": 130 |
| }, |
| { |
| "entropy": 0.7487856037914753, |
| "epoch": 0.6588235294117647, |
| "grad_norm": 0.2995753884315491, |
| "learning_rate": 7.58135693817893e-06, |
| "loss": 0.7511, |
| "mean_token_accuracy": 0.855742147564888, |
| "num_tokens": 251107.0, |
| "step": 140 |
| }, |
| { |
| "entropy": 0.7198293171823025, |
| "epoch": 0.7058823529411765, |
| "grad_norm": 0.2929564416408539, |
| "learning_rate": 7.226559750537462e-06, |
| "loss": 0.7237, |
| "mean_token_accuracy": 0.8601349666714668, |
| "num_tokens": 268729.0, |
| "step": 150 |
| }, |
| { |
| "epoch": 0.7058823529411765, |
| "eval_entropy": 0.7403583451112111, |
| "eval_loss": 0.698711633682251, |
| "eval_mean_token_accuracy": 0.8560926202932994, |
| "eval_num_tokens": 268729.0, |
| "eval_runtime": 19.698, |
| "eval_samples_per_second": 7.615, |
| "eval_steps_per_second": 7.615, |
| "step": 150 |
| }, |
| { |
| "entropy": 0.7665069714188576, |
| "epoch": 0.7529411764705882, |
| "grad_norm": 0.24647395312786102, |
| "learning_rate": 6.857173299431084e-06, |
| "loss": 0.7298, |
| "mean_token_accuracy": 0.8562039688229561, |
| "num_tokens": 286777.0, |
| "step": 160 |
| }, |
| { |
| "entropy": 0.6268041007220745, |
| "epoch": 0.8, |
| "grad_norm": 0.22048509120941162, |
| "learning_rate": 6.475617944903693e-06, |
| "loss": 0.6185, |
| "mean_token_accuracy": 0.8762901470065116, |
| "num_tokens": 304268.0, |
| "step": 170 |
| }, |
| { |
| "entropy": 0.7007844015955925, |
| "epoch": 0.8470588235294118, |
| "grad_norm": 0.2350086122751236, |
| "learning_rate": 6.084393782267039e-06, |
| "loss": 0.6875, |
| "mean_token_accuracy": 0.8667616620659828, |
| "num_tokens": 321505.0, |
| "step": 180 |
| }, |
| { |
| "entropy": 0.7066945165395737, |
| "epoch": 0.8941176470588236, |
| "grad_norm": 0.27258384227752686, |
| "learning_rate": 5.686064260528577e-06, |
| "loss": 0.7017, |
| "mean_token_accuracy": 0.8596053853631019, |
| "num_tokens": 340285.0, |
| "step": 190 |
| }, |
| { |
| "entropy": 0.7212481118738652, |
| "epoch": 0.9411764705882353, |
| "grad_norm": 0.25644779205322266, |
| "learning_rate": 5.283239385701882e-06, |
| "loss": 0.715, |
| "mean_token_accuracy": 0.8563221275806427, |
| "num_tokens": 360365.0, |
| "step": 200 |
| }, |
| { |
| "epoch": 0.9411764705882353, |
| "eval_entropy": 0.7153916182120641, |
| "eval_loss": 0.6829276084899902, |
| "eval_mean_token_accuracy": 0.8566614667574565, |
| "eval_num_tokens": 360365.0, |
| "eval_runtime": 19.6569, |
| "eval_samples_per_second": 7.631, |
| "eval_steps_per_second": 7.631, |
| "step": 200 |
| } |
| ], |
| "logging_steps": 10, |
| "max_steps": 400, |
| "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": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 1.655702451597312e+16, |
| "train_batch_size": 1, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
