| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.859154929577464, | |
| "eval_steps": 500, | |
| "global_step": 350, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.28169014084507044, | |
| "grad_norm": 6.4921875, | |
| "learning_rate": 0.00019959742939952392, | |
| "loss": 1.0479, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5633802816901409, | |
| "grad_norm": 7.4765625, | |
| "learning_rate": 0.00019839295885986296, | |
| "loss": 1.0329, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.8450704225352113, | |
| "grad_norm": 8.5546875, | |
| "learning_rate": 0.00019639628606958533, | |
| "loss": 0.9256, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.1267605633802817, | |
| "grad_norm": 7.9453125, | |
| "learning_rate": 0.00019362348706397373, | |
| "loss": 0.8794, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.408450704225352, | |
| "grad_norm": 9.7578125, | |
| "learning_rate": 0.0001900968867902419, | |
| "loss": 0.8596, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.6901408450704225, | |
| "grad_norm": 9.765625, | |
| "learning_rate": 0.00018584487936018661, | |
| "loss": 0.8522, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.971830985915493, | |
| "grad_norm": 10.0859375, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.8563, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.2535211267605635, | |
| "grad_norm": 8.421875, | |
| "learning_rate": 0.00017530714660036112, | |
| "loss": 0.826, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.535211267605634, | |
| "grad_norm": 9.7265625, | |
| "learning_rate": 0.00016910626489868649, | |
| "loss": 0.8266, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.816901408450704, | |
| "grad_norm": 11.2265625, | |
| "learning_rate": 0.00016234898018587337, | |
| "loss": 0.8513, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.0985915492957745, | |
| "grad_norm": 10.9765625, | |
| "learning_rate": 0.00015508969814521025, | |
| "loss": 0.8051, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 3.380281690140845, | |
| "grad_norm": 12.53125, | |
| "learning_rate": 0.00014738686624729986, | |
| "loss": 0.8422, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.6619718309859155, | |
| "grad_norm": 10.234375, | |
| "learning_rate": 0.00013930250316539238, | |
| "loss": 0.8595, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.943661971830986, | |
| "grad_norm": 12.8671875, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.8358, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 4.225352112676056, | |
| "grad_norm": 11.7109375, | |
| "learning_rate": 0.00012225209339563145, | |
| "loss": 0.8135, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 4.507042253521127, | |
| "grad_norm": 9.9296875, | |
| "learning_rate": 0.00011342332658176555, | |
| "loss": 0.7967, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.788732394366197, | |
| "grad_norm": 10.546875, | |
| "learning_rate": 0.00010448648303505151, | |
| "loss": 0.7877, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 5.070422535211268, | |
| "grad_norm": 14.2265625, | |
| "learning_rate": 9.551351696494854e-05, | |
| "loss": 0.7516, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 5.352112676056338, | |
| "grad_norm": 11.4375, | |
| "learning_rate": 8.657667341823448e-05, | |
| "loss": 0.7581, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 5.633802816901408, | |
| "grad_norm": 10.5859375, | |
| "learning_rate": 7.774790660436858e-05, | |
| "loss": 0.7491, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.915492957746479, | |
| "grad_norm": 9.6796875, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.7307, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 6.197183098591549, | |
| "grad_norm": 8.890625, | |
| "learning_rate": 6.069749683460765e-05, | |
| "loss": 0.7323, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 6.47887323943662, | |
| "grad_norm": 11.2734375, | |
| "learning_rate": 5.261313375270014e-05, | |
| "loss": 0.7145, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 6.76056338028169, | |
| "grad_norm": 10.4375, | |
| "learning_rate": 4.491030185478976e-05, | |
| "loss": 0.7525, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 7.042253521126761, | |
| "grad_norm": 10.9765625, | |
| "learning_rate": 3.7651019814126654e-05, | |
| "loss": 0.7138, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 7.323943661971831, | |
| "grad_norm": 9.625, | |
| "learning_rate": 3.089373510131354e-05, | |
| "loss": 0.7266, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 7.605633802816901, | |
| "grad_norm": 9.6484375, | |
| "learning_rate": 2.4692853399638917e-05, | |
| "loss": 0.7136, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 7.887323943661972, | |
| "grad_norm": 10.25, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.7194, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 8.169014084507042, | |
| "grad_norm": 11.3203125, | |
| "learning_rate": 1.415512063981339e-05, | |
| "loss": 0.7137, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 8.450704225352112, | |
| "grad_norm": 13.578125, | |
| "learning_rate": 9.903113209758096e-06, | |
| "loss": 0.7186, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 8.732394366197184, | |
| "grad_norm": 8.828125, | |
| "learning_rate": 6.37651293602628e-06, | |
| "loss": 0.7232, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 9.014084507042254, | |
| "grad_norm": 11.9140625, | |
| "learning_rate": 3.6037139304146762e-06, | |
| "loss": 0.7116, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 9.295774647887324, | |
| "grad_norm": 9.4453125, | |
| "learning_rate": 1.6070411401370334e-06, | |
| "loss": 0.7106, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 9.577464788732394, | |
| "grad_norm": 12.828125, | |
| "learning_rate": 4.025706004760932e-07, | |
| "loss": 0.727, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 9.859154929577464, | |
| "grad_norm": 10.4453125, | |
| "learning_rate": 0.0, | |
| "loss": 0.704, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 9.859154929577464, | |
| "step": 350, | |
| "total_flos": 2.2849945337856e+16, | |
| "train_loss": 0.7934087766919817, | |
| "train_runtime": 317.5625, | |
| "train_samples_per_second": 4.472, | |
| "train_steps_per_second": 1.102 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 350, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "total_flos": 2.2849945337856e+16, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |