| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.005108491590145719, | |
| "eval_steps": 9, | |
| "global_step": 100, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 5.1084915901457196e-05, | |
| "eval_loss": 3.4734065532684326, | |
| "eval_runtime": 182.7499, | |
| "eval_samples_per_second": 180.405, | |
| "eval_steps_per_second": 22.555, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.00015325474770437158, | |
| "grad_norm": 1.0560590028762817, | |
| "learning_rate": 1.5e-05, | |
| "loss": 3.4321, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.00030650949540874317, | |
| "grad_norm": 1.1394942998886108, | |
| "learning_rate": 3e-05, | |
| "loss": 3.4168, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.0004597642431131148, | |
| "grad_norm": 1.2241615056991577, | |
| "learning_rate": 4.5e-05, | |
| "loss": 3.366, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0004597642431131148, | |
| "eval_loss": 3.415942430496216, | |
| "eval_runtime": 183.8456, | |
| "eval_samples_per_second": 179.33, | |
| "eval_steps_per_second": 22.421, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.0006130189908174863, | |
| "grad_norm": 1.198073148727417, | |
| "learning_rate": 4.993910125649561e-05, | |
| "loss": 3.4657, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.000766273738521858, | |
| "grad_norm": 1.0100293159484863, | |
| "learning_rate": 4.962019382530521e-05, | |
| "loss": 3.2748, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0009195284862262296, | |
| "grad_norm": 0.9332877397537231, | |
| "learning_rate": 4.9031542398457974e-05, | |
| "loss": 3.246, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0009195284862262296, | |
| "eval_loss": 3.144437313079834, | |
| "eval_runtime": 184.0033, | |
| "eval_samples_per_second": 179.176, | |
| "eval_steps_per_second": 22.402, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.0010727832339306011, | |
| "grad_norm": 0.9414787292480469, | |
| "learning_rate": 4.817959636416969e-05, | |
| "loss": 3.0904, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 0.0012260379816349727, | |
| "grad_norm": 1.0662180185317993, | |
| "learning_rate": 4.707368982147318e-05, | |
| "loss": 2.9901, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 0.0013792927293393444, | |
| "grad_norm": 0.9672358632087708, | |
| "learning_rate": 4.572593931387604e-05, | |
| "loss": 3.0054, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.0013792927293393444, | |
| "eval_loss": 2.9016616344451904, | |
| "eval_runtime": 183.1925, | |
| "eval_samples_per_second": 179.969, | |
| "eval_steps_per_second": 22.501, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 0.001532547477043716, | |
| "grad_norm": 0.9545574188232422, | |
| "learning_rate": 4.415111107797445e-05, | |
| "loss": 2.804, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.0016858022247480875, | |
| "grad_norm": 0.8684502243995667, | |
| "learning_rate": 4.2366459261474933e-05, | |
| "loss": 2.7914, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 0.0018390569724524592, | |
| "grad_norm": 0.8820082545280457, | |
| "learning_rate": 4.039153688314145e-05, | |
| "loss": 2.7448, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0018390569724524592, | |
| "eval_loss": 2.6998445987701416, | |
| "eval_runtime": 182.5973, | |
| "eval_samples_per_second": 180.556, | |
| "eval_steps_per_second": 22.574, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 0.0019923117201568305, | |
| "grad_norm": 0.9635749459266663, | |
| "learning_rate": 3.824798160583012e-05, | |
| "loss": 2.6368, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 0.0021455664678612023, | |
| "grad_norm": 1.0027378797531128, | |
| "learning_rate": 3.5959278669726935e-05, | |
| "loss": 2.6553, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 0.002298821215565574, | |
| "grad_norm": 0.9840696454048157, | |
| "learning_rate": 3.355050358314172e-05, | |
| "loss": 2.6122, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.002298821215565574, | |
| "eval_loss": 2.5452194213867188, | |
| "eval_runtime": 182.8932, | |
| "eval_samples_per_second": 180.264, | |
| "eval_steps_per_second": 22.538, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.0024520759632699453, | |
| "grad_norm": 0.9413293600082397, | |
| "learning_rate": 3.104804738999169e-05, | |
| "loss": 2.5858, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 0.002605330710974317, | |
| "grad_norm": 1.006203293800354, | |
| "learning_rate": 2.8479327524001636e-05, | |
| "loss": 2.4787, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 0.002758585458678689, | |
| "grad_norm": 1.007853388786316, | |
| "learning_rate": 2.587248741756253e-05, | |
| "loss": 2.4619, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.002758585458678689, | |
| "eval_loss": 2.424272298812866, | |
| "eval_runtime": 182.9352, | |
| "eval_samples_per_second": 180.222, | |
| "eval_steps_per_second": 22.533, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 0.00291184020638306, | |
| "grad_norm": 1.1123751401901245, | |
| "learning_rate": 2.3256088156396868e-05, | |
| "loss": 2.3757, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 0.003065094954087432, | |
| "grad_norm": 1.0468323230743408, | |
| "learning_rate": 2.0658795558326743e-05, | |
| "loss": 2.3774, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.0032183497017918036, | |
| "grad_norm": 1.2713088989257812, | |
| "learning_rate": 1.8109066104575023e-05, | |
| "loss": 2.3676, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.0032183497017918036, | |
| "eval_loss": 2.315412759780884, | |
| "eval_runtime": 183.2875, | |
| "eval_samples_per_second": 179.876, | |
| "eval_steps_per_second": 22.489, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 0.003371604449496175, | |
| "grad_norm": 1.2722467184066772, | |
| "learning_rate": 1.56348351646022e-05, | |
| "loss": 2.3911, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 0.0035248591972005467, | |
| "grad_norm": 1.2411410808563232, | |
| "learning_rate": 1.3263210930352737e-05, | |
| "loss": 2.2346, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 0.0036781139449049184, | |
| "grad_norm": 1.144443392753601, | |
| "learning_rate": 1.1020177413231334e-05, | |
| "loss": 2.3083, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.0036781139449049184, | |
| "eval_loss": 2.2295360565185547, | |
| "eval_runtime": 182.7601, | |
| "eval_samples_per_second": 180.395, | |
| "eval_steps_per_second": 22.554, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 0.0038313686926092897, | |
| "grad_norm": 1.25135338306427, | |
| "learning_rate": 8.930309757836517e-06, | |
| "loss": 2.2523, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.003984623440313661, | |
| "grad_norm": 1.035020112991333, | |
| "learning_rate": 7.016504991533726e-06, | |
| "loss": 2.2171, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 0.004137878188018033, | |
| "grad_norm": 1.0950953960418701, | |
| "learning_rate": 5.299731159831953e-06, | |
| "loss": 2.1921, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.004137878188018033, | |
| "eval_loss": 2.1888010501861572, | |
| "eval_runtime": 182.9199, | |
| "eval_samples_per_second": 180.237, | |
| "eval_steps_per_second": 22.534, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 0.0042911329357224045, | |
| "grad_norm": 1.026548981666565, | |
| "learning_rate": 3.798797596089351e-06, | |
| "loss": 2.172, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 0.004444387683426776, | |
| "grad_norm": 0.9955553412437439, | |
| "learning_rate": 2.5301488425208296e-06, | |
| "loss": 2.1886, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 0.004597642431131148, | |
| "grad_norm": 1.0280815362930298, | |
| "learning_rate": 1.5076844803522922e-06, | |
| "loss": 2.1352, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.004597642431131148, | |
| "eval_loss": 2.1757705211639404, | |
| "eval_runtime": 182.8294, | |
| "eval_samples_per_second": 180.327, | |
| "eval_steps_per_second": 22.546, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.00475089717883552, | |
| "grad_norm": 1.1763354539871216, | |
| "learning_rate": 7.426068431000882e-07, | |
| "loss": 2.2136, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 0.004904151926539891, | |
| "grad_norm": 1.103149652481079, | |
| "learning_rate": 2.4329828146074095e-07, | |
| "loss": 2.2201, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 0.005057406674244262, | |
| "grad_norm": 1.0177910327911377, | |
| "learning_rate": 1.522932452260595e-08, | |
| "loss": 2.1461, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 0.005057406674244262, | |
| "eval_loss": 2.1734840869903564, | |
| "eval_runtime": 182.7199, | |
| "eval_samples_per_second": 180.435, | |
| "eval_steps_per_second": 22.559, | |
| "step": 99 | |
| } | |
| ], | |
| "logging_steps": 3, | |
| "max_steps": 100, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 9, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 2776751721676800.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |