| { | |
| "best_global_step": null, | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 9.654545454545454, | |
| "eval_steps": 500, | |
| "global_step": 270, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.03636363636363636, | |
| "grad_norm": 3.691458225250244, | |
| "learning_rate": 7.4074074074074075e-06, | |
| "loss": 2.6618, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.18181818181818182, | |
| "grad_norm": 3.446274518966675, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 2.6408, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 2.032472848892212, | |
| "learning_rate": 7.407407407407407e-05, | |
| "loss": 2.5317, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5454545454545454, | |
| "grad_norm": 0.9463324546813965, | |
| "learning_rate": 0.00011111111111111112, | |
| "loss": 2.36, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 0.7080094218254089, | |
| "learning_rate": 0.00014814814814814815, | |
| "loss": 2.2174, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.9090909090909091, | |
| "grad_norm": 0.5037975311279297, | |
| "learning_rate": 0.0001851851851851852, | |
| "loss": 2.1182, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_loss": 2.782132625579834, | |
| "eval_runtime": 0.8292, | |
| "eval_samples_per_second": 12.06, | |
| "eval_steps_per_second": 1.206, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.0727272727272728, | |
| "grad_norm": 0.41888633370399475, | |
| "learning_rate": 0.00019992479525042303, | |
| "loss": 2.0277, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.2545454545454544, | |
| "grad_norm": 0.27915704250335693, | |
| "learning_rate": 0.00019946562024066014, | |
| "loss": 1.9587, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.4363636363636363, | |
| "grad_norm": 0.20056034624576569, | |
| "learning_rate": 0.00019859096633447965, | |
| "loss": 1.9087, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.6181818181818182, | |
| "grad_norm": 0.16737522184848785, | |
| "learning_rate": 0.00019730448705798239, | |
| "loss": 1.8766, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.8, | |
| "grad_norm": 0.15048423409461975, | |
| "learning_rate": 0.00019561155617738797, | |
| "loss": 1.8481, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.981818181818182, | |
| "grad_norm": 0.1224176436662674, | |
| "learning_rate": 0.000193519245252219, | |
| "loss": 1.8354, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_loss": 2.737755537033081, | |
| "eval_runtime": 0.829, | |
| "eval_samples_per_second": 12.063, | |
| "eval_steps_per_second": 1.206, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.1454545454545455, | |
| "grad_norm": 0.1324545294046402, | |
| "learning_rate": 0.0001910362940966147, | |
| "loss": 1.8118, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.327272727272727, | |
| "grad_norm": 0.11611360311508179, | |
| "learning_rate": 0.0001881730742721608, | |
| "loss": 1.7937, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 2.509090909090909, | |
| "grad_norm": 0.1148991584777832, | |
| "learning_rate": 0.00018494154576472976, | |
| "loss": 1.7791, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 2.690909090909091, | |
| "grad_norm": 0.11438702791929245, | |
| "learning_rate": 0.00018135520702629675, | |
| "loss": 1.7654, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 2.8727272727272726, | |
| "grad_norm": 0.11716635525226593, | |
| "learning_rate": 0.00017742903859041325, | |
| "loss": 1.7604, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_loss": 2.7259373664855957, | |
| "eval_runtime": 0.8303, | |
| "eval_samples_per_second": 12.044, | |
| "eval_steps_per_second": 1.204, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 3.036363636363636, | |
| "grad_norm": 0.1302882581949234, | |
| "learning_rate": 0.00017317944049686124, | |
| "loss": 1.7453, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 3.2181818181818183, | |
| "grad_norm": 0.12489154189825058, | |
| "learning_rate": 0.0001686241637868734, | |
| "loss": 1.7396, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 3.4, | |
| "grad_norm": 0.10804688185453415, | |
| "learning_rate": 0.0001637822363550706, | |
| "loss": 1.7272, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 3.581818181818182, | |
| "grad_norm": 0.1448238343000412, | |
| "learning_rate": 0.0001586738834678418, | |
| "loss": 1.7231, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 3.7636363636363637, | |
| "grad_norm": 0.12403673678636551, | |
| "learning_rate": 0.00015332044328016914, | |
| "loss": 1.7101, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 3.9454545454545453, | |
| "grad_norm": 0.11520184576511383, | |
| "learning_rate": 0.0001477442777037949, | |
| "loss": 1.7035, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_loss": 2.724990129470825, | |
| "eval_runtime": 0.8296, | |
| "eval_samples_per_second": 12.053, | |
| "eval_steps_per_second": 1.205, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 4.109090909090909, | |
| "grad_norm": 0.11850611865520477, | |
| "learning_rate": 0.0001419686789990429, | |
| "loss": 1.6998, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 4.290909090909091, | |
| "grad_norm": 0.141310453414917, | |
| "learning_rate": 0.00013601777248047105, | |
| "loss": 1.6942, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 4.472727272727273, | |
| "grad_norm": 0.14388997852802277, | |
| "learning_rate": 0.00012991641574276418, | |
| "loss": 1.6887, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 4.654545454545454, | |
| "grad_norm": 0.11356977373361588, | |
| "learning_rate": 0.00012369009482781192, | |
| "loss": 1.6845, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 4.836363636363636, | |
| "grad_norm": 0.13505423069000244, | |
| "learning_rate": 0.00011736481776669306, | |
| "loss": 1.6801, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "grad_norm": 0.18071481585502625, | |
| "learning_rate": 0.00011096700594125318, | |
| "loss": 1.6822, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_loss": 2.7262730598449707, | |
| "eval_runtime": 0.8327, | |
| "eval_samples_per_second": 12.009, | |
| "eval_steps_per_second": 1.201, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 5.181818181818182, | |
| "grad_norm": 0.12405228614807129, | |
| "learning_rate": 0.00010452338371907064, | |
| "loss": 1.671, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 5.363636363636363, | |
| "grad_norm": 0.15709254145622253, | |
| "learning_rate": 9.806086682281758e-05, | |
| "loss": 1.6697, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 5.545454545454545, | |
| "grad_norm": 0.1405353993177414, | |
| "learning_rate": 9.160644990030931e-05, | |
| "loss": 1.6707, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 5.7272727272727275, | |
| "grad_norm": 0.13487176597118378, | |
| "learning_rate": 8.518709376487515e-05, | |
| "loss": 1.6619, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 5.909090909090909, | |
| "grad_norm": 0.12394227087497711, | |
| "learning_rate": 7.882961277705895e-05, | |
| "loss": 1.6619, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_loss": 2.7253997325897217, | |
| "eval_runtime": 0.8321, | |
| "eval_samples_per_second": 12.017, | |
| "eval_steps_per_second": 1.202, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 6.072727272727272, | |
| "grad_norm": 0.11816684156656265, | |
| "learning_rate": 7.256056283806986e-05, | |
| "loss": 1.6573, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 6.254545454545455, | |
| "grad_norm": 0.14117498695850372, | |
| "learning_rate": 6.640613046284581e-05, | |
| "loss": 1.6622, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 6.4363636363636365, | |
| "grad_norm": 0.1342514008283615, | |
| "learning_rate": 6.039202339608432e-05, | |
| "loss": 1.6535, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 6.618181818181818, | |
| "grad_norm": 0.13483189046382904, | |
| "learning_rate": 5.4543363228149946e-05, | |
| "loss": 1.6532, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 6.8, | |
| "grad_norm": 0.1636153757572174, | |
| "learning_rate": 4.888458045941269e-05, | |
| "loss": 1.6482, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 6.9818181818181815, | |
| "grad_norm": 0.1563912183046341, | |
| "learning_rate": 4.343931245134616e-05, | |
| "loss": 1.6471, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_loss": 2.7240517139434814, | |
| "eval_runtime": 0.8312, | |
| "eval_samples_per_second": 12.031, | |
| "eval_steps_per_second": 1.203, | |
| "step": 196 | |
| }, | |
| { | |
| "epoch": 7.1454545454545455, | |
| "grad_norm": 0.11320989578962326, | |
| "learning_rate": 3.8230304690654304e-05, | |
| "loss": 1.6472, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 7.327272727272727, | |
| "grad_norm": 0.111383818089962, | |
| "learning_rate": 3.3279315778858036e-05, | |
| "loss": 1.6488, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 7.509090909090909, | |
| "grad_norm": 0.10844731330871582, | |
| "learning_rate": 2.8607026544210114e-05, | |
| "loss": 1.6458, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 7.690909090909091, | |
| "grad_norm": 0.10823339223861694, | |
| "learning_rate": 2.423295365558821e-05, | |
| "loss": 1.6456, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 7.872727272727273, | |
| "grad_norm": 0.10790830850601196, | |
| "learning_rate": 2.01753680992107e-05, | |
| "loss": 1.6458, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 2.7252650260925293, | |
| "eval_runtime": 0.8302, | |
| "eval_samples_per_second": 12.045, | |
| "eval_steps_per_second": 1.204, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 8.036363636363637, | |
| "grad_norm": 0.11462420970201492, | |
| "learning_rate": 1.6451218858706374e-05, | |
| "loss": 1.643, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 8.218181818181819, | |
| "grad_norm": 0.10164881497621536, | |
| "learning_rate": 1.307606211733522e-05, | |
| "loss": 1.6435, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 8.4, | |
| "grad_norm": 0.11715802550315857, | |
| "learning_rate": 1.0063996278090704e-05, | |
| "loss": 1.6436, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 8.581818181818182, | |
| "grad_norm": 0.1077931597828865, | |
| "learning_rate": 7.427603073110967e-06, | |
| "loss": 1.6437, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 8.763636363636364, | |
| "grad_norm": 0.09881118685007095, | |
| "learning_rate": 5.177895008392353e-06, | |
| "loss": 1.6415, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 8.945454545454545, | |
| "grad_norm": 0.0973580852150917, | |
| "learning_rate": 3.3242693633337983e-06, | |
| "loss": 1.641, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_loss": 2.725569009780884, | |
| "eval_runtime": 0.8306, | |
| "eval_samples_per_second": 12.039, | |
| "eval_steps_per_second": 1.204, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 9.10909090909091, | |
| "grad_norm": 0.10264136642217636, | |
| "learning_rate": 1.874468937261531e-06, | |
| "loss": 1.6464, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 9.290909090909091, | |
| "grad_norm": 0.1021399274468422, | |
| "learning_rate": 8.345497068998897e-07, | |
| "loss": 1.6443, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 9.472727272727273, | |
| "grad_norm": 0.10423731058835983, | |
| "learning_rate": 2.088555298867978e-07, | |
| "loss": 1.6436, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 9.654545454545454, | |
| "grad_norm": 0.09860274940729141, | |
| "learning_rate": 0.0, | |
| "loss": 1.6383, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 9.654545454545454, | |
| "eval_loss": 2.725593328475952, | |
| "eval_runtime": 0.8317, | |
| "eval_samples_per_second": 12.024, | |
| "eval_steps_per_second": 1.202, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 9.654545454545454, | |
| "step": 270, | |
| "total_flos": 8.156088875152835e+17, | |
| "train_loss": 1.7710220513520418, | |
| "train_runtime": 1245.0854, | |
| "train_samples_per_second": 112.233, | |
| "train_steps_per_second": 0.217 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 270, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 100, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 8.156088875152835e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |