| { | |
| "best_metric": 0.819804310798645, | |
| "best_model_checkpoint": "data/Mistral-7B_task-1_180-samples_config-2_full_auto/checkpoint-42", | |
| "epoch": 12.0, | |
| "eval_steps": 500, | |
| "global_step": 102, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.11764705882352941, | |
| "grad_norm": 2.005297899246216, | |
| "learning_rate": 2.5e-06, | |
| "loss": 1.7613, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.23529411764705882, | |
| "grad_norm": 2.0482828617095947, | |
| "learning_rate": 5e-06, | |
| "loss": 1.711, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.47058823529411764, | |
| "grad_norm": 2.018578052520752, | |
| "learning_rate": 1e-05, | |
| "loss": 1.7421, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.7058823529411765, | |
| "grad_norm": 2.0357279777526855, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.7221, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "grad_norm": 1.6246639490127563, | |
| "learning_rate": 2e-05, | |
| "loss": 1.6633, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.9411764705882353, | |
| "eval_loss": 1.5707345008850098, | |
| "eval_runtime": 14.9986, | |
| "eval_samples_per_second": 2.4, | |
| "eval_steps_per_second": 2.4, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 1.1764705882352942, | |
| "grad_norm": 1.2116508483886719, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.5474, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.4117647058823528, | |
| "grad_norm": 1.048179268836975, | |
| "learning_rate": 3e-05, | |
| "loss": 1.4427, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.6470588235294117, | |
| "grad_norm": 1.1144566535949707, | |
| "learning_rate": 3.5e-05, | |
| "loss": 1.3752, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 1.8823529411764706, | |
| "grad_norm": 1.3725205659866333, | |
| "learning_rate": 4e-05, | |
| "loss": 1.2689, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 1.126222014427185, | |
| "eval_runtime": 14.968, | |
| "eval_samples_per_second": 2.405, | |
| "eval_steps_per_second": 2.405, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 2.1176470588235294, | |
| "grad_norm": 1.1397361755371094, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.1517, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 2.3529411764705883, | |
| "grad_norm": 1.017498254776001, | |
| "learning_rate": 5e-05, | |
| "loss": 1.0251, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.588235294117647, | |
| "grad_norm": 0.9621643424034119, | |
| "learning_rate": 5.500000000000001e-05, | |
| "loss": 0.9395, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 2.8235294117647056, | |
| "grad_norm": 0.8679986000061035, | |
| "learning_rate": 6e-05, | |
| "loss": 0.9431, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 2.9411764705882355, | |
| "eval_loss": 0.8983803987503052, | |
| "eval_runtime": 14.966, | |
| "eval_samples_per_second": 2.405, | |
| "eval_steps_per_second": 2.405, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 3.0588235294117645, | |
| "grad_norm": 0.7188401818275452, | |
| "learning_rate": 6.500000000000001e-05, | |
| "loss": 0.8541, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 3.2941176470588234, | |
| "grad_norm": 0.5674646496772766, | |
| "learning_rate": 7e-05, | |
| "loss": 0.8561, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 3.5294117647058822, | |
| "grad_norm": 0.5988804697990417, | |
| "learning_rate": 7.500000000000001e-05, | |
| "loss": 0.8433, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 3.764705882352941, | |
| "grad_norm": 0.5540823340415955, | |
| "learning_rate": 8e-05, | |
| "loss": 0.8285, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 0.5860567688941956, | |
| "learning_rate": 8.5e-05, | |
| "loss": 0.7827, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 0.842862069606781, | |
| "eval_runtime": 14.9617, | |
| "eval_samples_per_second": 2.406, | |
| "eval_steps_per_second": 2.406, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 4.235294117647059, | |
| "grad_norm": 0.5514520406723022, | |
| "learning_rate": 9e-05, | |
| "loss": 0.7561, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 4.470588235294118, | |
| "grad_norm": 0.6599115133285522, | |
| "learning_rate": 9.5e-05, | |
| "loss": 0.7423, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 4.705882352941177, | |
| "grad_norm": 0.6465198993682861, | |
| "learning_rate": 0.0001, | |
| "loss": 0.7392, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 4.9411764705882355, | |
| "grad_norm": 0.6529595255851746, | |
| "learning_rate": 9.999238475781957e-05, | |
| "loss": 0.7185, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 4.9411764705882355, | |
| "eval_loss": 0.819804310798645, | |
| "eval_runtime": 14.9729, | |
| "eval_samples_per_second": 2.404, | |
| "eval_steps_per_second": 2.404, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 5.176470588235294, | |
| "grad_norm": 0.7368524670600891, | |
| "learning_rate": 9.99695413509548e-05, | |
| "loss": 0.6768, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 5.411764705882353, | |
| "grad_norm": 0.6277140378952026, | |
| "learning_rate": 9.99314767377287e-05, | |
| "loss": 0.6157, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 5.647058823529412, | |
| "grad_norm": 0.8047894239425659, | |
| "learning_rate": 9.987820251299122e-05, | |
| "loss": 0.5943, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 5.882352941176471, | |
| "grad_norm": 0.8778942227363586, | |
| "learning_rate": 9.980973490458728e-05, | |
| "loss": 0.6317, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 0.8236051201820374, | |
| "eval_runtime": 14.9598, | |
| "eval_samples_per_second": 2.406, | |
| "eval_steps_per_second": 2.406, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 6.117647058823529, | |
| "grad_norm": 0.779517412185669, | |
| "learning_rate": 9.972609476841367e-05, | |
| "loss": 0.5526, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 6.352941176470588, | |
| "grad_norm": 0.8343177437782288, | |
| "learning_rate": 9.962730758206611e-05, | |
| "loss": 0.5131, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 6.588235294117647, | |
| "grad_norm": 0.8886534571647644, | |
| "learning_rate": 9.951340343707852e-05, | |
| "loss": 0.482, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 6.823529411764706, | |
| "grad_norm": 1.1148805618286133, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 0.4646, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 6.9411764705882355, | |
| "eval_loss": 0.885874330997467, | |
| "eval_runtime": 14.97, | |
| "eval_samples_per_second": 2.405, | |
| "eval_steps_per_second": 2.405, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 7.0588235294117645, | |
| "grad_norm": 0.9171388149261475, | |
| "learning_rate": 9.924038765061042e-05, | |
| "loss": 0.4544, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 7.294117647058823, | |
| "grad_norm": 0.9372969269752502, | |
| "learning_rate": 9.908135917238321e-05, | |
| "loss": 0.3585, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 7.529411764705882, | |
| "grad_norm": 1.2094287872314453, | |
| "learning_rate": 9.890738003669029e-05, | |
| "loss": 0.3365, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 7.764705882352941, | |
| "grad_norm": 1.365533709526062, | |
| "learning_rate": 9.871850323926177e-05, | |
| "loss": 0.3193, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 1.1985735893249512, | |
| "learning_rate": 9.851478631379982e-05, | |
| "loss": 0.3485, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.9817339777946472, | |
| "eval_runtime": 14.9708, | |
| "eval_samples_per_second": 2.405, | |
| "eval_steps_per_second": 2.405, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 8.235294117647058, | |
| "grad_norm": 0.995296061038971, | |
| "learning_rate": 9.829629131445342e-05, | |
| "loss": 0.2496, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 8.470588235294118, | |
| "grad_norm": 1.2525056600570679, | |
| "learning_rate": 9.806308479691595e-05, | |
| "loss": 0.2441, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 8.705882352941176, | |
| "grad_norm": 1.290474534034729, | |
| "learning_rate": 9.781523779815179e-05, | |
| "loss": 0.2304, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 8.941176470588236, | |
| "grad_norm": 1.9896327257156372, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 0.2504, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 8.941176470588236, | |
| "eval_loss": 1.1164960861206055, | |
| "eval_runtime": 14.9584, | |
| "eval_samples_per_second": 2.407, | |
| "eval_steps_per_second": 2.407, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 9.176470588235293, | |
| "grad_norm": 1.183082938194275, | |
| "learning_rate": 9.727592877996585e-05, | |
| "loss": 0.1912, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 9.411764705882353, | |
| "grad_norm": 1.0704456567764282, | |
| "learning_rate": 9.698463103929542e-05, | |
| "loss": 0.1715, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 9.647058823529411, | |
| "grad_norm": 1.1364245414733887, | |
| "learning_rate": 9.667902132486009e-05, | |
| "loss": 0.1662, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 9.882352941176471, | |
| "grad_norm": 1.416142225265503, | |
| "learning_rate": 9.635919272833938e-05, | |
| "loss": 0.1848, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_loss": 1.1628649234771729, | |
| "eval_runtime": 14.9583, | |
| "eval_samples_per_second": 2.407, | |
| "eval_steps_per_second": 2.407, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 10.117647058823529, | |
| "grad_norm": 1.0128824710845947, | |
| "learning_rate": 9.602524267262203e-05, | |
| "loss": 0.1609, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 10.352941176470589, | |
| "grad_norm": 1.0560998916625977, | |
| "learning_rate": 9.567727288213005e-05, | |
| "loss": 0.1361, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 10.588235294117647, | |
| "grad_norm": 1.2621757984161377, | |
| "learning_rate": 9.53153893518325e-05, | |
| "loss": 0.1336, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 10.823529411764707, | |
| "grad_norm": 1.1911786794662476, | |
| "learning_rate": 9.493970231495835e-05, | |
| "loss": 0.1505, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 10.941176470588236, | |
| "eval_loss": 1.238175392150879, | |
| "eval_runtime": 14.972, | |
| "eval_samples_per_second": 2.404, | |
| "eval_steps_per_second": 2.404, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 11.058823529411764, | |
| "grad_norm": 0.889838695526123, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 0.1318, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 11.294117647058824, | |
| "grad_norm": 1.1177611351013184, | |
| "learning_rate": 9.414737964294636e-05, | |
| "loss": 0.1168, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 11.529411764705882, | |
| "grad_norm": 0.7929391264915466, | |
| "learning_rate": 9.373098535696979e-05, | |
| "loss": 0.1153, | |
| "step": 98 | |
| }, | |
| { | |
| "epoch": 11.764705882352942, | |
| "grad_norm": 1.2638691663742065, | |
| "learning_rate": 9.330127018922194e-05, | |
| "loss": 0.1208, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 1.1541279554367065, | |
| "learning_rate": 9.285836503510562e-05, | |
| "loss": 0.1227, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_loss": 1.2854546308517456, | |
| "eval_runtime": 14.9759, | |
| "eval_samples_per_second": 2.404, | |
| "eval_steps_per_second": 2.404, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "step": 102, | |
| "total_flos": 2.747429741710541e+16, | |
| "train_loss": 0.645151332459029, | |
| "train_runtime": 1987.9039, | |
| "train_samples_per_second": 3.421, | |
| "train_steps_per_second": 0.201 | |
| } | |
| ], | |
| "logging_steps": 2, | |
| "max_steps": 400, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 50, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "EarlyStoppingCallback": { | |
| "args": { | |
| "early_stopping_patience": 7, | |
| "early_stopping_threshold": 0.0 | |
| }, | |
| "attributes": { | |
| "early_stopping_patience_counter": 0 | |
| } | |
| }, | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.747429741710541e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |