| { | |
| "best_metric": 1.5659970045089722, | |
| "best_model_checkpoint": "finetuned-websites/checkpoint-200", | |
| "epoch": 10.0, | |
| "eval_steps": 50, | |
| "global_step": 240, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4166666666666667, | |
| "grad_norm": 4.125311851501465, | |
| "learning_rate": 1.0000000000000002e-06, | |
| "loss": 1.9272, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.8333333333333334, | |
| "grad_norm": 3.0228281021118164, | |
| "learning_rate": 2.0000000000000003e-06, | |
| "loss": 1.928, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.25, | |
| "grad_norm": 2.5070629119873047, | |
| "learning_rate": 3e-06, | |
| "loss": 1.921, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.6666666666666665, | |
| "grad_norm": 2.7787866592407227, | |
| "learning_rate": 4.000000000000001e-06, | |
| "loss": 1.8972, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.0833333333333335, | |
| "grad_norm": 2.7862040996551514, | |
| "learning_rate": 5e-06, | |
| "loss": 1.8678, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.0833333333333335, | |
| "eval_accuracy": 0.3088235294117647, | |
| "eval_loss": 1.8965561389923096, | |
| "eval_runtime": 2.6013, | |
| "eval_samples_per_second": 26.141, | |
| "eval_steps_per_second": 3.46, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.5, | |
| "grad_norm": 3.540709972381592, | |
| "learning_rate": 6e-06, | |
| "loss": 1.8526, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.9166666666666665, | |
| "grad_norm": 3.426781415939331, | |
| "learning_rate": 7.000000000000001e-06, | |
| "loss": 1.8198, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.3333333333333335, | |
| "grad_norm": 4.007580757141113, | |
| "learning_rate": 8.000000000000001e-06, | |
| "loss": 1.7369, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.75, | |
| "grad_norm": 4.467859745025635, | |
| "learning_rate": 9e-06, | |
| "loss": 1.7144, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.166666666666667, | |
| "grad_norm": 3.9431729316711426, | |
| "learning_rate": 1e-05, | |
| "loss": 1.6483, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.166666666666667, | |
| "eval_accuracy": 0.3382352941176471, | |
| "eval_loss": 1.793816089630127, | |
| "eval_runtime": 2.9764, | |
| "eval_samples_per_second": 22.847, | |
| "eval_steps_per_second": 3.024, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.583333333333333, | |
| "grad_norm": 3.2608771324157715, | |
| "learning_rate": 1.1000000000000001e-05, | |
| "loss": 1.5856, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 3.638972520828247, | |
| "learning_rate": 1.2e-05, | |
| "loss": 1.5157, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 5.416666666666667, | |
| "grad_norm": 3.3878164291381836, | |
| "learning_rate": 1.3000000000000001e-05, | |
| "loss": 1.4388, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 5.833333333333333, | |
| "grad_norm": 4.086641311645508, | |
| "learning_rate": 1.4000000000000001e-05, | |
| "loss": 1.4522, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "grad_norm": 3.5550928115844727, | |
| "learning_rate": 1.5e-05, | |
| "loss": 1.3419, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 6.25, | |
| "eval_accuracy": 0.38235294117647056, | |
| "eval_loss": 1.6614487171173096, | |
| "eval_runtime": 2.1852, | |
| "eval_samples_per_second": 31.118, | |
| "eval_steps_per_second": 4.119, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 4.774816036224365, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 1.3147, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 7.083333333333333, | |
| "grad_norm": 4.092309951782227, | |
| "learning_rate": 1.7000000000000003e-05, | |
| "loss": 1.2643, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 7.5, | |
| "grad_norm": 4.992920398712158, | |
| "learning_rate": 1.8e-05, | |
| "loss": 1.1924, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 7.916666666666667, | |
| "grad_norm": 6.251681327819824, | |
| "learning_rate": 1.9e-05, | |
| "loss": 1.1461, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 8.333333333333334, | |
| "grad_norm": 5.500473499298096, | |
| "learning_rate": 2e-05, | |
| "loss": 1.1189, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 8.333333333333334, | |
| "eval_accuracy": 0.39705882352941174, | |
| "eval_loss": 1.5659970045089722, | |
| "eval_runtime": 2.0786, | |
| "eval_samples_per_second": 32.715, | |
| "eval_steps_per_second": 4.33, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 8.75, | |
| "grad_norm": 4.436168670654297, | |
| "learning_rate": 2.1e-05, | |
| "loss": 1.0734, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 9.166666666666666, | |
| "grad_norm": 4.706101417541504, | |
| "learning_rate": 2.2000000000000003e-05, | |
| "loss": 0.9833, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 9.583333333333334, | |
| "grad_norm": 4.173801898956299, | |
| "learning_rate": 2.3000000000000003e-05, | |
| "loss": 0.9902, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "grad_norm": 5.312504291534424, | |
| "learning_rate": 2.4e-05, | |
| "loss": 0.9339, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "step": 240, | |
| "total_flos": 2.9603266596900864e+17, | |
| "train_loss": 1.4860339760780334, | |
| "train_runtime": 184.3796, | |
| "train_samples_per_second": 20.718, | |
| "train_steps_per_second": 1.302 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 240, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 50, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2.9603266596900864e+17, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |