| { |
| "best_global_step": null, |
| "best_metric": null, |
| "best_model_checkpoint": null, |
| "epoch": 2.0, |
| "eval_steps": 1000, |
| "global_step": 11030, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.00018132366273798732, |
| "grad_norm": 7.9868855476379395, |
| "learning_rate": 0.0, |
| "loss": 3.5226, |
| "step": 1 |
| }, |
| { |
| "epoch": 0.1813236627379873, |
| "grad_norm": 6.11737060546875, |
| "learning_rate": 7.2443799854967376e-06, |
| "loss": 2.9981, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.1813236627379873, |
| "eval_cosine_accuracy": 0.9449994564056396, |
| "eval_loss": 0.5478581190109253, |
| "eval_runtime": 34.6287, |
| "eval_samples_per_second": 274.599, |
| "eval_steps_per_second": 2.166, |
| "step": 1000 |
| }, |
| { |
| "epoch": 0.3626473254759746, |
| "grad_norm": 6.67965841293335, |
| "learning_rate": 9.763322644678577e-06, |
| "loss": 2.3124, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.3626473254759746, |
| "eval_cosine_accuracy": 0.9545693397521973, |
| "eval_loss": 0.495151549577713, |
| "eval_runtime": 34.4914, |
| "eval_samples_per_second": 275.692, |
| "eval_steps_per_second": 2.174, |
| "step": 2000 |
| }, |
| { |
| "epoch": 0.543970988213962, |
| "grad_norm": 7.58078145980835, |
| "learning_rate": 9.381966712475189e-06, |
| "loss": 1.974, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.543970988213962, |
| "eval_cosine_accuracy": 0.9580397605895996, |
| "eval_loss": 0.4687798023223877, |
| "eval_runtime": 35.422, |
| "eval_samples_per_second": 268.449, |
| "eval_steps_per_second": 2.117, |
| "step": 3000 |
| }, |
| { |
| "epoch": 0.7252946509519492, |
| "grad_norm": 11.277643203735352, |
| "learning_rate": 9.000229042601925e-06, |
| "loss": 1.4538, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.7252946509519492, |
| "eval_cosine_accuracy": 0.9532022476196289, |
| "eval_loss": 0.4629732668399811, |
| "eval_runtime": 35.494, |
| "eval_samples_per_second": 267.905, |
| "eval_steps_per_second": 2.113, |
| "step": 4000 |
| }, |
| { |
| "epoch": 0.9066183136899365, |
| "grad_norm": 9.13899040222168, |
| "learning_rate": 8.618873110398535e-06, |
| "loss": 1.0996, |
| "step": 5000 |
| }, |
| { |
| "epoch": 0.9066183136899365, |
| "eval_cosine_accuracy": 0.9505731463432312, |
| "eval_loss": 0.4556906819343567, |
| "eval_runtime": 35.5731, |
| "eval_samples_per_second": 267.309, |
| "eval_steps_per_second": 2.108, |
| "step": 5000 |
| }, |
| { |
| "epoch": 1.087941976427924, |
| "grad_norm": 6.146666049957275, |
| "learning_rate": 8.237135440525271e-06, |
| "loss": 1.3439, |
| "step": 6000 |
| }, |
| { |
| "epoch": 1.087941976427924, |
| "eval_cosine_accuracy": 0.9636133909225464, |
| "eval_loss": 0.42725491523742676, |
| "eval_runtime": 36.4213, |
| "eval_samples_per_second": 261.084, |
| "eval_steps_per_second": 2.059, |
| "step": 6000 |
| }, |
| { |
| "epoch": 1.2692656391659112, |
| "grad_norm": 6.063191890716553, |
| "learning_rate": 7.855397770652008e-06, |
| "loss": 1.6419, |
| "step": 7000 |
| }, |
| { |
| "epoch": 1.2692656391659112, |
| "eval_cosine_accuracy": 0.9639289379119873, |
| "eval_loss": 0.4292740523815155, |
| "eval_runtime": 35.7914, |
| "eval_samples_per_second": 265.678, |
| "eval_steps_per_second": 2.095, |
| "step": 7000 |
| }, |
| { |
| "epoch": 1.4505893019038985, |
| "grad_norm": 6.539251327514648, |
| "learning_rate": 7.473660100778746e-06, |
| "loss": 1.5472, |
| "step": 8000 |
| }, |
| { |
| "epoch": 1.4505893019038985, |
| "eval_cosine_accuracy": 0.9652960300445557, |
| "eval_loss": 0.43033671379089355, |
| "eval_runtime": 35.6592, |
| "eval_samples_per_second": 266.664, |
| "eval_steps_per_second": 2.103, |
| "step": 8000 |
| }, |
| { |
| "epoch": 1.6319129646418857, |
| "grad_norm": 7.785158157348633, |
| "learning_rate": 7.092685906245229e-06, |
| "loss": 1.3452, |
| "step": 9000 |
| }, |
| { |
| "epoch": 1.6319129646418857, |
| "eval_cosine_accuracy": 0.9624566435813904, |
| "eval_loss": 0.425350546836853, |
| "eval_runtime": 35.8844, |
| "eval_samples_per_second": 264.99, |
| "eval_steps_per_second": 2.09, |
| "step": 9000 |
| }, |
| { |
| "epoch": 1.8132366273798732, |
| "grad_norm": 10.031052589416504, |
| "learning_rate": 6.710948236371966e-06, |
| "loss": 0.9822, |
| "step": 10000 |
| }, |
| { |
| "epoch": 1.8132366273798732, |
| "eval_cosine_accuracy": 0.9575139284133911, |
| "eval_loss": 0.42917805910110474, |
| "eval_runtime": 34.9605, |
| "eval_samples_per_second": 271.993, |
| "eval_steps_per_second": 2.145, |
| "step": 10000 |
| }, |
| { |
| "epoch": 1.9945602901178603, |
| "grad_norm": 10.17534351348877, |
| "learning_rate": 6.329210566498702e-06, |
| "loss": 0.8215, |
| "step": 11000 |
| }, |
| { |
| "epoch": 1.9945602901178603, |
| "eval_cosine_accuracy": 0.9543590545654297, |
| "eval_loss": 0.4252549707889557, |
| "eval_runtime": 34.9403, |
| "eval_samples_per_second": 272.15, |
| "eval_steps_per_second": 2.147, |
| "step": 11000 |
| } |
| ], |
| "logging_steps": 1000, |
| "max_steps": 27575, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 5, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": false |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 0.0, |
| "train_batch_size": 128, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
