| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 3.568065810991625, | |
| "eval_steps": 500, | |
| "global_step": 9000, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.19822587838842362, | |
| "grad_norm": 47.1356201171875, | |
| "learning_rate": 0.0000024777006937561945, | |
| "loss": 21.4029, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.19822587838842362, | |
| "eval_loss": 586.7496948242188, | |
| "eval_runtime": 573.4198, | |
| "eval_samples_per_second": 53, | |
| "eval_steps_per_second": 6.625, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 0.39645175677684724, | |
| "grad_norm": 50.89448165893555, | |
| "learning_rate": 0.000004955401387512389, | |
| "loss": 11.7925, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.39645175677684724, | |
| "eval_loss": 584.6571044921875, | |
| "eval_runtime": 571.5589, | |
| "eval_samples_per_second": 53.172, | |
| "eval_steps_per_second": 6.647, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 0.5946776351652708, | |
| "grad_norm": 56.098140716552734, | |
| "learning_rate": 0.000007433102081268584, | |
| "loss": 9.7243, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.5946776351652708, | |
| "eval_loss": 438.975830078125, | |
| "eval_runtime": 571.006, | |
| "eval_samples_per_second": 53.224, | |
| "eval_steps_per_second": 6.653, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 0.7929035135536945, | |
| "grad_norm": 49.88700485229492, | |
| "learning_rate": 0.000009910802775024778, | |
| "loss": 8.6629, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.7929035135536945, | |
| "eval_loss": 443.8860778808594, | |
| "eval_runtime": 569.8536, | |
| "eval_samples_per_second": 53.331, | |
| "eval_steps_per_second": 6.667, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 0.991129391942118, | |
| "grad_norm": 37.3896484375, | |
| "learning_rate": 0.00000991223686038578, | |
| "loss": 7.7527, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 0.991129391942118, | |
| "eval_loss": 359.1586608886719, | |
| "eval_runtime": 570.3107, | |
| "eval_samples_per_second": 53.289, | |
| "eval_steps_per_second": 6.661, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 1.1893552703305417, | |
| "grad_norm": 45.05537414550781, | |
| "learning_rate": 0.000009639072360499928, | |
| "loss": 7.0532, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.1893552703305417, | |
| "eval_loss": 297.4908447265625, | |
| "eval_runtime": 573.7765, | |
| "eval_samples_per_second": 52.967, | |
| "eval_steps_per_second": 6.621, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 1.3875811487189653, | |
| "grad_norm": 42.01622772216797, | |
| "learning_rate": 0.000009190700553732768, | |
| "loss": 6.8798, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.3875811487189653, | |
| "eval_loss": 272.1216125488281, | |
| "eval_runtime": 570.5906, | |
| "eval_samples_per_second": 53.262, | |
| "eval_steps_per_second": 6.658, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 1.5858070271073887, | |
| "grad_norm": 26.336191177368164, | |
| "learning_rate": 0.000008584055432624292, | |
| "loss": 6.5736, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.5858070271073887, | |
| "eval_loss": 279.0899353027344, | |
| "eval_runtime": 578.6986, | |
| "eval_samples_per_second": 52.516, | |
| "eval_steps_per_second": 6.565, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 1.7840329054958124, | |
| "grad_norm": 35.772953033447266, | |
| "learning_rate": 0.00000784204861605915, | |
| "loss": 6.3985, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.7840329054958124, | |
| "eval_loss": 221.53627014160156, | |
| "eval_runtime": 569.7883, | |
| "eval_samples_per_second": 53.337, | |
| "eval_steps_per_second": 6.667, | |
| "step": 4500 | |
| }, | |
| { | |
| "epoch": 1.982258783884236, | |
| "grad_norm": 34.2270622253418, | |
| "learning_rate": 0.000006992704029062722, | |
| "loss": 6.0866, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 1.982258783884236, | |
| "eval_loss": 190.98533630371094, | |
| "eval_runtime": 571.0123, | |
| "eval_samples_per_second": 53.223, | |
| "eval_steps_per_second": 6.653, | |
| "step": 5000 | |
| }, | |
| { | |
| "epoch": 2.1804846622726597, | |
| "grad_norm": 34.992889404296875, | |
| "learning_rate": 0.000006068099502317051, | |
| "loss": 5.732, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.1804846622726597, | |
| "eval_loss": 225.56539916992188, | |
| "eval_runtime": 570.6646, | |
| "eval_samples_per_second": 53.255, | |
| "eval_steps_per_second": 6.657, | |
| "step": 5500 | |
| }, | |
| { | |
| "epoch": 2.3787105406610833, | |
| "grad_norm": 35.05290222167969, | |
| "learning_rate": 0.000005103155264797376, | |
| "loss": 5.8384, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.3787105406610833, | |
| "eval_loss": 206.36277770996094, | |
| "eval_runtime": 569.1115, | |
| "eval_samples_per_second": 53.401, | |
| "eval_steps_per_second": 6.675, | |
| "step": 6000 | |
| }, | |
| { | |
| "epoch": 2.576936419049507, | |
| "grad_norm": 38.268341064453125, | |
| "learning_rate": 0.0000041343150854317275, | |
| "loss": 5.6189, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.576936419049507, | |
| "eval_loss": 211.91796875, | |
| "eval_runtime": 569.7007, | |
| "eval_samples_per_second": 53.346, | |
| "eval_steps_per_second": 6.668, | |
| "step": 6500 | |
| }, | |
| { | |
| "epoch": 2.7751622974379306, | |
| "grad_norm": 42.83818054199219, | |
| "learning_rate": 0.0000031981698740904465, | |
| "loss": 5.5445, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.7751622974379306, | |
| "eval_loss": 199.4452362060547, | |
| "eval_runtime": 570.9073, | |
| "eval_samples_per_second": 53.233, | |
| "eval_steps_per_second": 6.654, | |
| "step": 7000 | |
| }, | |
| { | |
| "epoch": 2.9733881758263543, | |
| "grad_norm": 42.24338150024414, | |
| "learning_rate": 0.0000023300757253939837, | |
| "loss": 5.546, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 2.9733881758263543, | |
| "eval_loss": 190.01214599609375, | |
| "eval_runtime": 571.9686, | |
| "eval_samples_per_second": 53.134, | |
| "eval_steps_per_second": 6.642, | |
| "step": 7500 | |
| }, | |
| { | |
| "epoch": 3.171614054214778, | |
| "grad_norm": 34.228065490722656, | |
| "learning_rate": 0.0000015628185987094685, | |
| "loss": 5.3668, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.171614054214778, | |
| "eval_loss": 194.59649658203125, | |
| "eval_runtime": 571.773, | |
| "eval_samples_per_second": 53.152, | |
| "eval_steps_per_second": 6.644, | |
| "step": 8000 | |
| }, | |
| { | |
| "epoch": 3.3698399326032016, | |
| "grad_norm": 30.128808975219727, | |
| "learning_rate": 9.253760663628896e-7, | |
| "loss": 5.338, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.3698399326032016, | |
| "eval_loss": 173.66778564453125, | |
| "eval_runtime": 569.7472, | |
| "eval_samples_per_second": 53.341, | |
| "eval_steps_per_second": 6.668, | |
| "step": 8500 | |
| }, | |
| { | |
| "epoch": 3.568065810991625, | |
| "grad_norm": 54.32496643066406, | |
| "learning_rate": 4.4182289604595254e-7, | |
| "loss": 5.2961, | |
| "step": 9000 | |
| }, | |
| { | |
| "epoch": 3.568065810991625, | |
| "eval_loss": 184.6847381591797, | |
| "eval_runtime": 571.6945, | |
| "eval_samples_per_second": 53.16, | |
| "eval_steps_per_second": 6.645, | |
| "step": 9000 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 10088, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 4, | |
| "save_steps": 1000, | |
| "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, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } |