| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.0, | |
| "global_step": 7044, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.43, | |
| "learning_rate": 4.2901760363429874e-05, | |
| "loss": 3.9512, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.43, | |
| "eval_accuracy": 0.6988671808767591, | |
| "eval_loss": 1.3428400754928589, | |
| "eval_runtime": 171.0039, | |
| "eval_samples_per_second": 292.929, | |
| "eval_steps_per_second": 4.579, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "learning_rate": 3.580352072685974e-05, | |
| "loss": 1.0044, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.85, | |
| "eval_accuracy": 0.9013550464459579, | |
| "eval_loss": 0.5994584560394287, | |
| "eval_runtime": 170.4181, | |
| "eval_samples_per_second": 293.936, | |
| "eval_steps_per_second": 4.595, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "learning_rate": 2.8705281090289608e-05, | |
| "loss": 0.5483, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.28, | |
| "eval_accuracy": 0.9309371495565297, | |
| "eval_loss": 0.446216881275177, | |
| "eval_runtime": 171.3996, | |
| "eval_samples_per_second": 292.253, | |
| "eval_steps_per_second": 4.568, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "learning_rate": 2.160704145371948e-05, | |
| "loss": 0.436, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.7, | |
| "eval_accuracy": 0.9376707432277173, | |
| "eval_loss": 0.40857475996017456, | |
| "eval_runtime": 171.9437, | |
| "eval_samples_per_second": 291.328, | |
| "eval_steps_per_second": 4.554, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 1.4508801817149347e-05, | |
| "loss": 0.3764, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "eval_accuracy": 0.9462951194712761, | |
| "eval_loss": 0.3566935658454895, | |
| "eval_runtime": 171.3127, | |
| "eval_samples_per_second": 292.401, | |
| "eval_steps_per_second": 4.571, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "learning_rate": 7.410562180579217e-06, | |
| "loss": 0.337, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.56, | |
| "eval_accuracy": 0.9496153650211422, | |
| "eval_loss": 0.3342791795730591, | |
| "eval_runtime": 171.713, | |
| "eval_samples_per_second": 291.719, | |
| "eval_steps_per_second": 4.56, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 3.1232254400908575e-07, | |
| "loss": 0.3115, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "eval_accuracy": 0.95252108208063, | |
| "eval_loss": 0.30891212821006775, | |
| "eval_runtime": 171.2387, | |
| "eval_samples_per_second": 292.527, | |
| "eval_steps_per_second": 4.573, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "step": 7044, | |
| "total_flos": 1.2593673024207894e+16, | |
| "train_loss": 0.9907198370619974, | |
| "train_runtime": 6301.0396, | |
| "train_samples_per_second": 71.534, | |
| "train_steps_per_second": 1.118 | |
| } | |
| ], | |
| "max_steps": 7044, | |
| "num_train_epochs": 3, | |
| "total_flos": 1.2593673024207894e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |