| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.02021018593371059, | |
| "eval_steps": 5, | |
| "global_step": 50, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.0004042037186742118, | |
| "grad_norm": 1.263440728187561, | |
| "learning_rate": 1e-05, | |
| "loss": 4.0397, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0004042037186742118, | |
| "eval_loss": 1.917946219444275, | |
| "eval_runtime": 452.0131, | |
| "eval_samples_per_second": 4.61, | |
| "eval_steps_per_second": 0.577, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0008084074373484236, | |
| "grad_norm": 1.0014753341674805, | |
| "learning_rate": 2e-05, | |
| "loss": 3.5685, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.0012126111560226355, | |
| "grad_norm": 0.9920203685760498, | |
| "learning_rate": 3e-05, | |
| "loss": 3.4815, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.0016168148746968471, | |
| "grad_norm": 0.9647438526153564, | |
| "learning_rate": 4e-05, | |
| "loss": 3.5763, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.002021018593371059, | |
| "grad_norm": 0.9799220561981201, | |
| "learning_rate": 5e-05, | |
| "loss": 3.5288, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.002021018593371059, | |
| "eval_loss": 1.8881479501724243, | |
| "eval_runtime": 452.9863, | |
| "eval_samples_per_second": 4.601, | |
| "eval_steps_per_second": 0.576, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.002425222312045271, | |
| "grad_norm": 1.4729738235473633, | |
| "learning_rate": 6e-05, | |
| "loss": 3.8284, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0028294260307194824, | |
| "grad_norm": 1.3433350324630737, | |
| "learning_rate": 7e-05, | |
| "loss": 3.8095, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.0032336297493936943, | |
| "grad_norm": 1.1986558437347412, | |
| "learning_rate": 8e-05, | |
| "loss": 3.8635, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.003637833468067906, | |
| "grad_norm": 1.181183934211731, | |
| "learning_rate": 9e-05, | |
| "loss": 3.4141, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.004042037186742118, | |
| "grad_norm": 1.2625378370285034, | |
| "learning_rate": 0.0001, | |
| "loss": 3.2466, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.004042037186742118, | |
| "eval_loss": 1.6723251342773438, | |
| "eval_runtime": 453.0563, | |
| "eval_samples_per_second": 4.6, | |
| "eval_steps_per_second": 0.576, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00444624090541633, | |
| "grad_norm": 1.3650546073913574, | |
| "learning_rate": 9.98458666866564e-05, | |
| "loss": 3.4225, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.004850444624090542, | |
| "grad_norm": 1.01077401638031, | |
| "learning_rate": 9.938441702975689e-05, | |
| "loss": 3.3738, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.005254648342764754, | |
| "grad_norm": 2.3417210578918457, | |
| "learning_rate": 9.861849601988383e-05, | |
| "loss": 3.4139, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.005658852061438965, | |
| "grad_norm": 1.6219801902770996, | |
| "learning_rate": 9.755282581475769e-05, | |
| "loss": 2.7161, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.006063055780113177, | |
| "grad_norm": 1.0307327508926392, | |
| "learning_rate": 9.619397662556435e-05, | |
| "loss": 2.8197, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.006063055780113177, | |
| "eval_loss": 1.4082988500595093, | |
| "eval_runtime": 453.1209, | |
| "eval_samples_per_second": 4.599, | |
| "eval_steps_per_second": 0.576, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0064672594987873885, | |
| "grad_norm": 1.0622483491897583, | |
| "learning_rate": 9.45503262094184e-05, | |
| "loss": 3.0642, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.0068714632174616, | |
| "grad_norm": 0.7077268362045288, | |
| "learning_rate": 9.263200821770461e-05, | |
| "loss": 2.7863, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.007275666936135812, | |
| "grad_norm": 0.8749211430549622, | |
| "learning_rate": 9.045084971874738e-05, | |
| "loss": 2.7657, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.007679870654810024, | |
| "grad_norm": 0.8615348935127258, | |
| "learning_rate": 8.802029828000156e-05, | |
| "loss": 2.6338, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 0.008084074373484237, | |
| "grad_norm": 0.8538162708282471, | |
| "learning_rate": 8.535533905932738e-05, | |
| "loss": 2.7195, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.008084074373484237, | |
| "eval_loss": 1.3105684518814087, | |
| "eval_runtime": 453.3652, | |
| "eval_samples_per_second": 4.597, | |
| "eval_steps_per_second": 0.576, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.008488278092158448, | |
| "grad_norm": 0.7278123497962952, | |
| "learning_rate": 8.247240241650918e-05, | |
| "loss": 2.6497, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.00889248181083266, | |
| "grad_norm": 0.6712276935577393, | |
| "learning_rate": 7.938926261462366e-05, | |
| "loss": 2.6245, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 0.009296685529506871, | |
| "grad_norm": 0.5658913254737854, | |
| "learning_rate": 7.612492823579745e-05, | |
| "loss": 2.724, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 0.009700889248181084, | |
| "grad_norm": 0.7421402931213379, | |
| "learning_rate": 7.269952498697734e-05, | |
| "loss": 2.2569, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.010105092966855295, | |
| "grad_norm": 1.5091322660446167, | |
| "learning_rate": 6.91341716182545e-05, | |
| "loss": 2.7908, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.010105092966855295, | |
| "eval_loss": 1.2647299766540527, | |
| "eval_runtime": 453.2429, | |
| "eval_samples_per_second": 4.598, | |
| "eval_steps_per_second": 0.576, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.010509296685529508, | |
| "grad_norm": 0.93433678150177, | |
| "learning_rate": 6.545084971874738e-05, | |
| "loss": 2.4645, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 0.010913500404203719, | |
| "grad_norm": 0.8287851810455322, | |
| "learning_rate": 6.167226819279528e-05, | |
| "loss": 2.6179, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.01131770412287793, | |
| "grad_norm": 0.8977319598197937, | |
| "learning_rate": 5.782172325201155e-05, | |
| "loss": 2.7171, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 0.011721907841552142, | |
| "grad_norm": 0.7713293433189392, | |
| "learning_rate": 5.392295478639225e-05, | |
| "loss": 2.6567, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 0.012126111560226353, | |
| "grad_norm": 0.7393156886100769, | |
| "learning_rate": 5e-05, | |
| "loss": 2.5136, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.012126111560226353, | |
| "eval_loss": 1.2371978759765625, | |
| "eval_runtime": 453.0604, | |
| "eval_samples_per_second": 4.6, | |
| "eval_steps_per_second": 0.576, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.012530315278900566, | |
| "grad_norm": 0.6830640435218811, | |
| "learning_rate": 4.607704521360776e-05, | |
| "loss": 2.6676, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 0.012934518997574777, | |
| "grad_norm": 0.7319803237915039, | |
| "learning_rate": 4.2178276747988446e-05, | |
| "loss": 2.3856, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 0.01333872271624899, | |
| "grad_norm": 0.7554514408111572, | |
| "learning_rate": 3.832773180720475e-05, | |
| "loss": 2.5566, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0137429264349232, | |
| "grad_norm": 0.6606221795082092, | |
| "learning_rate": 3.4549150281252636e-05, | |
| "loss": 2.3467, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 0.014147130153597413, | |
| "grad_norm": 0.6036873459815979, | |
| "learning_rate": 3.086582838174551e-05, | |
| "loss": 2.0719, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.014147130153597413, | |
| "eval_loss": 1.224223017692566, | |
| "eval_runtime": 453.2044, | |
| "eval_samples_per_second": 4.598, | |
| "eval_steps_per_second": 0.576, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.014551333872271624, | |
| "grad_norm": 0.9149614572525024, | |
| "learning_rate": 2.7300475013022663e-05, | |
| "loss": 2.2007, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.014955537590945837, | |
| "grad_norm": 0.47336137294769287, | |
| "learning_rate": 2.3875071764202563e-05, | |
| "loss": 2.4277, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 0.015359741309620048, | |
| "grad_norm": 0.6594354510307312, | |
| "learning_rate": 2.061073738537635e-05, | |
| "loss": 2.8188, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 0.01576394502829426, | |
| "grad_norm": 0.6771719455718994, | |
| "learning_rate": 1.7527597583490822e-05, | |
| "loss": 2.3827, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.016168148746968473, | |
| "grad_norm": 0.7665150761604309, | |
| "learning_rate": 1.4644660940672627e-05, | |
| "loss": 2.6651, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.016168148746968473, | |
| "eval_loss": 1.2175811529159546, | |
| "eval_runtime": 453.0444, | |
| "eval_samples_per_second": 4.6, | |
| "eval_steps_per_second": 0.576, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.016572352465642683, | |
| "grad_norm": 1.0414700508117676, | |
| "learning_rate": 1.1979701719998453e-05, | |
| "loss": 2.9944, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 0.016976556184316895, | |
| "grad_norm": 0.7823891043663025, | |
| "learning_rate": 9.549150281252633e-06, | |
| "loss": 2.3473, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.017380759902991108, | |
| "grad_norm": 0.49085983633995056, | |
| "learning_rate": 7.367991782295391e-06, | |
| "loss": 2.2603, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 0.01778496362166532, | |
| "grad_norm": 0.7287192344665527, | |
| "learning_rate": 5.449673790581611e-06, | |
| "loss": 2.5286, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 0.01818916734033953, | |
| "grad_norm": 0.5006673336029053, | |
| "learning_rate": 3.8060233744356633e-06, | |
| "loss": 2.0819, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.01818916734033953, | |
| "eval_loss": 1.215002417564392, | |
| "eval_runtime": 453.0566, | |
| "eval_samples_per_second": 4.6, | |
| "eval_steps_per_second": 0.576, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.018593371059013743, | |
| "grad_norm": 0.6527900099754333, | |
| "learning_rate": 2.4471741852423237e-06, | |
| "loss": 2.2517, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 0.018997574777687955, | |
| "grad_norm": 1.0630096197128296, | |
| "learning_rate": 1.3815039801161721e-06, | |
| "loss": 2.9461, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 0.019401778496362168, | |
| "grad_norm": 0.761471152305603, | |
| "learning_rate": 6.15582970243117e-07, | |
| "loss": 2.2641, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.019805982215036377, | |
| "grad_norm": 0.7035945653915405, | |
| "learning_rate": 1.5413331334360182e-07, | |
| "loss": 2.2215, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 0.02021018593371059, | |
| "grad_norm": 0.7908831238746643, | |
| "learning_rate": 0.0, | |
| "loss": 2.6591, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.02021018593371059, | |
| "eval_loss": 1.2144694328308105, | |
| "eval_runtime": 453.5575, | |
| "eval_samples_per_second": 4.595, | |
| "eval_steps_per_second": 0.575, | |
| "step": 50 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 50, | |
| "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": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 6.538264861212672e+16, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |