| { |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 1.0, |
| "eval_steps": 250, |
| "global_step": 1351, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.050333086602516654, |
| "grad_norm": 4.236826419830322, |
| "learning_rate": 5e-07, |
| "loss": 0.4956, |
| "step": 68 |
| }, |
| { |
| "epoch": 0.10066617320503331, |
| "grad_norm": 3.3058083057403564, |
| "learning_rate": 1e-06, |
| "loss": 0.5465, |
| "step": 136 |
| }, |
| { |
| "epoch": 0.15099925980754997, |
| "grad_norm": 3.0439271926879883, |
| "learning_rate": 9.440329218106997e-07, |
| "loss": 0.4505, |
| "step": 204 |
| }, |
| { |
| "epoch": 0.1850481125092524, |
| "eval_sft_sg_values_eval_loss": 0.3208533227443695, |
| "eval_sft_sg_values_eval_runtime": 28.8285, |
| "eval_sft_sg_values_eval_samples_per_second": 17.344, |
| "eval_sft_sg_values_eval_steps_per_second": 8.672, |
| "step": 250 |
| }, |
| { |
| "epoch": 0.20133234641006661, |
| "grad_norm": 3.065812110900879, |
| "learning_rate": 8.880658436213991e-07, |
| "loss": 0.4528, |
| "step": 272 |
| }, |
| { |
| "epoch": 0.25166543301258326, |
| "grad_norm": 3.464308977127075, |
| "learning_rate": 8.320987654320987e-07, |
| "loss": 0.3473, |
| "step": 340 |
| }, |
| { |
| "epoch": 0.30199851961509994, |
| "grad_norm": 1.8523364067077637, |
| "learning_rate": 7.761316872427983e-07, |
| "loss": 0.395, |
| "step": 408 |
| }, |
| { |
| "epoch": 0.35233160621761656, |
| "grad_norm": 1.4938207864761353, |
| "learning_rate": 7.201646090534979e-07, |
| "loss": 0.3465, |
| "step": 476 |
| }, |
| { |
| "epoch": 0.3700962250185048, |
| "eval_sft_sg_values_eval_loss": 0.20385459065437317, |
| "eval_sft_sg_values_eval_runtime": 28.85, |
| "eval_sft_sg_values_eval_samples_per_second": 17.331, |
| "eval_sft_sg_values_eval_steps_per_second": 8.666, |
| "step": 500 |
| }, |
| { |
| "epoch": 0.40266469282013323, |
| "grad_norm": 2.14213490486145, |
| "learning_rate": 6.641975308641975e-07, |
| "loss": 0.2638, |
| "step": 544 |
| }, |
| { |
| "epoch": 0.4529977794226499, |
| "grad_norm": 1.6940499544143677, |
| "learning_rate": 6.082304526748971e-07, |
| "loss": 0.3279, |
| "step": 612 |
| }, |
| { |
| "epoch": 0.5033308660251665, |
| "grad_norm": 1.800094485282898, |
| "learning_rate": 5.522633744855967e-07, |
| "loss": 0.2363, |
| "step": 680 |
| }, |
| { |
| "epoch": 0.5536639526276832, |
| "grad_norm": 1.591727614402771, |
| "learning_rate": 4.962962962962963e-07, |
| "loss": 0.2179, |
| "step": 748 |
| }, |
| { |
| "epoch": 0.5551443375277573, |
| "eval_sft_sg_values_eval_loss": 0.17694725096225739, |
| "eval_sft_sg_values_eval_runtime": 28.885, |
| "eval_sft_sg_values_eval_samples_per_second": 17.31, |
| "eval_sft_sg_values_eval_steps_per_second": 8.655, |
| "step": 750 |
| }, |
| { |
| "epoch": 0.6039970392301999, |
| "grad_norm": 4.029638290405273, |
| "learning_rate": 4.403292181069959e-07, |
| "loss": 0.2013, |
| "step": 816 |
| }, |
| { |
| "epoch": 0.6543301258327165, |
| "grad_norm": 1.7355984449386597, |
| "learning_rate": 3.8436213991769547e-07, |
| "loss": 0.2014, |
| "step": 884 |
| }, |
| { |
| "epoch": 0.7046632124352331, |
| "grad_norm": 2.6946229934692383, |
| "learning_rate": 3.2839506172839506e-07, |
| "loss": 0.2954, |
| "step": 952 |
| }, |
| { |
| "epoch": 0.7401924500370096, |
| "eval_sft_sg_values_eval_loss": 0.16258862614631653, |
| "eval_sft_sg_values_eval_runtime": 28.8506, |
| "eval_sft_sg_values_eval_samples_per_second": 17.331, |
| "eval_sft_sg_values_eval_steps_per_second": 8.665, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.7549962990377498, |
| "grad_norm": 2.2922027111053467, |
| "learning_rate": 2.7242798353909465e-07, |
| "loss": 0.2395, |
| "step": 1020 |
| }, |
| { |
| "epoch": 0.8053293856402665, |
| "grad_norm": 1.3803716897964478, |
| "learning_rate": 2.1646090534979421e-07, |
| "loss": 0.249, |
| "step": 1088 |
| }, |
| { |
| "epoch": 0.8556624722427831, |
| "grad_norm": 1.8623427152633667, |
| "learning_rate": 1.604938271604938e-07, |
| "loss": 0.2043, |
| "step": 1156 |
| }, |
| { |
| "epoch": 0.9059955588452998, |
| "grad_norm": 2.6236300468444824, |
| "learning_rate": 1.0452674897119341e-07, |
| "loss": 0.2347, |
| "step": 1224 |
| }, |
| { |
| "epoch": 0.9252405625462621, |
| "eval_sft_sg_values_eval_loss": 0.15717695653438568, |
| "eval_sft_sg_values_eval_runtime": 28.8569, |
| "eval_sft_sg_values_eval_samples_per_second": 17.327, |
| "eval_sft_sg_values_eval_steps_per_second": 8.663, |
| "step": 1250 |
| }, |
| { |
| "epoch": 0.9563286454478165, |
| "grad_norm": 1.8058063983917236, |
| "learning_rate": 4.8559670781893e-08, |
| "loss": 0.27, |
| "step": 1292 |
| }, |
| { |
| "epoch": 1.0, |
| "step": 1351, |
| "total_flos": 9.434186513311334e+16, |
| "train_loss": 0.3092916518295367, |
| "train_runtime": 2173.1432, |
| "train_samples_per_second": 4.973, |
| "train_steps_per_second": 0.622 |
| } |
| ], |
| "logging_steps": 68, |
| "max_steps": 1351, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 1, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 9.434186513311334e+16, |
| "train_batch_size": 2, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
