| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.5952380952380952, | |
| "eval_steps": 25, | |
| "global_step": 475, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03132832080200501, | |
| "grad_norm": 3.161522626876831, | |
| "learning_rate": 2.3797595190380762e-05, | |
| "loss": 2.5111, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.03132832080200501, | |
| "eval_loss": 2.4335546493530273, | |
| "eval_runtime": 722.1607, | |
| "eval_samples_per_second": 2.209, | |
| "eval_steps_per_second": 0.277, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.06265664160401002, | |
| "grad_norm": 6.950015068054199, | |
| "learning_rate": 2.2545090180360722e-05, | |
| "loss": 2.3676, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.06265664160401002, | |
| "eval_loss": 2.241159439086914, | |
| "eval_runtime": 724.4073, | |
| "eval_samples_per_second": 2.202, | |
| "eval_steps_per_second": 0.276, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.09398496240601503, | |
| "grad_norm": 8.31242847442627, | |
| "learning_rate": 2.1292585170340683e-05, | |
| "loss": 2.2132, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.09398496240601503, | |
| "eval_loss": 2.1175358295440674, | |
| "eval_runtime": 720.5313, | |
| "eval_samples_per_second": 2.214, | |
| "eval_steps_per_second": 0.278, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.12531328320802004, | |
| "grad_norm": 11.33967399597168, | |
| "learning_rate": 2.0040080160320643e-05, | |
| "loss": 2.1464, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.12531328320802004, | |
| "eval_loss": 2.0354082584381104, | |
| "eval_runtime": 721.3533, | |
| "eval_samples_per_second": 2.211, | |
| "eval_steps_per_second": 0.277, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.15664160401002505, | |
| "grad_norm": 6.970797061920166, | |
| "learning_rate": 1.87875751503006e-05, | |
| "loss": 2.0109, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.15664160401002505, | |
| "eval_loss": 1.9767951965332031, | |
| "eval_runtime": 720.7941, | |
| "eval_samples_per_second": 2.213, | |
| "eval_steps_per_second": 0.277, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.18796992481203006, | |
| "grad_norm": 15.953432083129883, | |
| "learning_rate": 1.7535070140280564e-05, | |
| "loss": 2.0434, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.18796992481203006, | |
| "eval_loss": 1.9400724172592163, | |
| "eval_runtime": 725.3731, | |
| "eval_samples_per_second": 2.199, | |
| "eval_steps_per_second": 0.276, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.21929824561403508, | |
| "grad_norm": 9.164532661437988, | |
| "learning_rate": 1.628256513026052e-05, | |
| "loss": 1.9165, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.21929824561403508, | |
| "eval_loss": 1.916611671447754, | |
| "eval_runtime": 721.2615, | |
| "eval_samples_per_second": 2.211, | |
| "eval_steps_per_second": 0.277, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.2506265664160401, | |
| "grad_norm": 11.897910118103027, | |
| "learning_rate": 1.5030060120240483e-05, | |
| "loss": 2.0037, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2506265664160401, | |
| "eval_loss": 1.8966965675354004, | |
| "eval_runtime": 721.0344, | |
| "eval_samples_per_second": 2.212, | |
| "eval_steps_per_second": 0.277, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.2819548872180451, | |
| "grad_norm": 16.23155403137207, | |
| "learning_rate": 1.3777555110220442e-05, | |
| "loss": 1.9693, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.2819548872180451, | |
| "eval_loss": 1.8825273513793945, | |
| "eval_runtime": 720.7907, | |
| "eval_samples_per_second": 2.213, | |
| "eval_steps_per_second": 0.277, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.3132832080200501, | |
| "grad_norm": 11.5318021774292, | |
| "learning_rate": 1.25250501002004e-05, | |
| "loss": 1.8635, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.3132832080200501, | |
| "eval_loss": 1.8641014099121094, | |
| "eval_runtime": 720.4196, | |
| "eval_samples_per_second": 2.214, | |
| "eval_steps_per_second": 0.278, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.34461152882205515, | |
| "grad_norm": 16.45474624633789, | |
| "learning_rate": 1.1272545090180361e-05, | |
| "loss": 1.8685, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.34461152882205515, | |
| "eval_loss": 1.852317452430725, | |
| "eval_runtime": 720.5223, | |
| "eval_samples_per_second": 2.214, | |
| "eval_steps_per_second": 0.278, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.37593984962406013, | |
| "grad_norm": 17.96300506591797, | |
| "learning_rate": 1.0020040080160322e-05, | |
| "loss": 1.7522, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.37593984962406013, | |
| "eval_loss": 1.8399490118026733, | |
| "eval_runtime": 720.4464, | |
| "eval_samples_per_second": 2.214, | |
| "eval_steps_per_second": 0.278, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.40726817042606517, | |
| "grad_norm": 13.310808181762695, | |
| "learning_rate": 8.767535070140282e-06, | |
| "loss": 1.8459, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.40726817042606517, | |
| "eval_loss": 1.8312922716140747, | |
| "eval_runtime": 720.395, | |
| "eval_samples_per_second": 2.214, | |
| "eval_steps_per_second": 0.278, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.43859649122807015, | |
| "grad_norm": 13.073380470275879, | |
| "learning_rate": 7.515030060120242e-06, | |
| "loss": 1.8879, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.43859649122807015, | |
| "eval_loss": 1.8224186897277832, | |
| "eval_runtime": 720.0905, | |
| "eval_samples_per_second": 2.215, | |
| "eval_steps_per_second": 0.278, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.4699248120300752, | |
| "grad_norm": 15.554895401000977, | |
| "learning_rate": 6.2625250501002e-06, | |
| "loss": 1.9055, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.4699248120300752, | |
| "eval_loss": 1.8179148435592651, | |
| "eval_runtime": 721.2094, | |
| "eval_samples_per_second": 2.212, | |
| "eval_steps_per_second": 0.277, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.5012531328320802, | |
| "grad_norm": 15.814194679260254, | |
| "learning_rate": 5.010020040080161e-06, | |
| "loss": 1.8801, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.5012531328320802, | |
| "eval_loss": 1.8140182495117188, | |
| "eval_runtime": 721.3168, | |
| "eval_samples_per_second": 2.211, | |
| "eval_steps_per_second": 0.277, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.5325814536340853, | |
| "grad_norm": 9.506321907043457, | |
| "learning_rate": 3.757515030060121e-06, | |
| "loss": 1.8526, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.5325814536340853, | |
| "eval_loss": 1.8105257749557495, | |
| "eval_runtime": 720.746, | |
| "eval_samples_per_second": 2.213, | |
| "eval_steps_per_second": 0.277, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 0.5639097744360902, | |
| "grad_norm": 16.16132926940918, | |
| "learning_rate": 2.5050100200400804e-06, | |
| "loss": 1.89, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.5639097744360902, | |
| "eval_loss": 1.8077690601348877, | |
| "eval_runtime": 721.7445, | |
| "eval_samples_per_second": 2.21, | |
| "eval_steps_per_second": 0.277, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 0.5952380952380952, | |
| "grad_norm": 15.308736801147461, | |
| "learning_rate": 1.2525050100200402e-06, | |
| "loss": 1.8555, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 0.5952380952380952, | |
| "eval_loss": 1.8060969114303589, | |
| "eval_runtime": 722.1684, | |
| "eval_samples_per_second": 2.209, | |
| "eval_steps_per_second": 0.277, | |
| "step": 475 | |
| } | |
| ], | |
| "logging_steps": 25, | |
| "max_steps": 500, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 25, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": false | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1489311152517120.0, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |