| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 2.979683972911964, | |
| "eval_steps": 50, | |
| "global_step": 330, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.045146726862302484, | |
| "grad_norm": 18.47902498006474, | |
| "learning_rate": 5e-07, | |
| "loss": 1.7294, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.09029345372460497, | |
| "grad_norm": 12.268601512539195, | |
| "learning_rate": 1e-06, | |
| "loss": 1.5978, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.13544018058690746, | |
| "grad_norm": 7.473453913673138, | |
| "learning_rate": 9.993977281025862e-07, | |
| "loss": 1.2909, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.18058690744920994, | |
| "grad_norm": 4.242889467689194, | |
| "learning_rate": 9.975923633360984e-07, | |
| "loss": 1.1281, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.22573363431151242, | |
| "grad_norm": 3.7511187742202794, | |
| "learning_rate": 9.945882549823904e-07, | |
| "loss": 1.0669, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.2708803611738149, | |
| "grad_norm": 3.7686497891232653, | |
| "learning_rate": 9.90392640201615e-07, | |
| "loss": 1.0131, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3160270880361174, | |
| "grad_norm": 3.62119206079501, | |
| "learning_rate": 9.85015626597272e-07, | |
| "loss": 1.0054, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.3611738148984199, | |
| "grad_norm": 3.849448662188167, | |
| "learning_rate": 9.784701678661044e-07, | |
| "loss": 0.9772, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.40632054176072235, | |
| "grad_norm": 3.717637360909183, | |
| "learning_rate": 9.707720325915103e-07, | |
| "loss": 0.9666, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.45146726862302483, | |
| "grad_norm": 3.335114517852652, | |
| "learning_rate": 9.619397662556433e-07, | |
| "loss": 0.9349, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.45146726862302483, | |
| "eval_loss": 0.9357019066810608, | |
| "eval_runtime": 56.3673, | |
| "eval_samples_per_second": 55.883, | |
| "eval_steps_per_second": 0.887, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.4966139954853273, | |
| "grad_norm": 3.5546589218606477, | |
| "learning_rate": 9.519946465617217e-07, | |
| "loss": 0.9345, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.5417607223476298, | |
| "grad_norm": 3.531001152680907, | |
| "learning_rate": 9.409606321741774e-07, | |
| "loss": 0.9188, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5869074492099323, | |
| "grad_norm": 3.5692359406823804, | |
| "learning_rate": 9.28864305000136e-07, | |
| "loss": 0.9116, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.6320541760722348, | |
| "grad_norm": 3.702385519615215, | |
| "learning_rate": 9.157348061512726e-07, | |
| "loss": 0.9161, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6772009029345373, | |
| "grad_norm": 3.4298013712048716, | |
| "learning_rate": 9.016037657403223e-07, | |
| "loss": 0.8958, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.7223476297968398, | |
| "grad_norm": 3.4040303578876725, | |
| "learning_rate": 8.865052266813685e-07, | |
| "loss": 0.8658, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.7674943566591422, | |
| "grad_norm": 3.5133332427335895, | |
| "learning_rate": 8.704755626774795e-07, | |
| "loss": 0.8939, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.8126410835214447, | |
| "grad_norm": 3.500121545145948, | |
| "learning_rate": 8.535533905932737e-07, | |
| "loss": 0.8927, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8577878103837472, | |
| "grad_norm": 3.2225864446847545, | |
| "learning_rate": 8.357794774235092e-07, | |
| "loss": 0.8711, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.9029345372460497, | |
| "grad_norm": 3.632794695134748, | |
| "learning_rate": 8.171966420818227e-07, | |
| "loss": 0.892, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.9029345372460497, | |
| "eval_loss": 0.8807913064956665, | |
| "eval_runtime": 55.4559, | |
| "eval_samples_per_second": 56.802, | |
| "eval_steps_per_second": 0.902, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.9480812641083521, | |
| "grad_norm": 3.8825573929061155, | |
| "learning_rate": 7.978496522462167e-07, | |
| "loss": 0.8746, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.9932279909706546, | |
| "grad_norm": 3.325010031368483, | |
| "learning_rate": 7.777851165098011e-07, | |
| "loss": 0.8682, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.0383747178329572, | |
| "grad_norm": 3.349437996562217, | |
| "learning_rate": 7.570513720966107e-07, | |
| "loss": 0.8283, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.0835214446952597, | |
| "grad_norm": 3.5602949032422853, | |
| "learning_rate": 7.356983684129989e-07, | |
| "loss": 0.8152, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.1286681715575622, | |
| "grad_norm": 3.4642871655890644, | |
| "learning_rate": 7.13777546715141e-07, | |
| "loss": 0.8002, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.1738148984198646, | |
| "grad_norm": 3.4004037196199204, | |
| "learning_rate": 6.913417161825449e-07, | |
| "loss": 0.7893, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.2189616252821671, | |
| "grad_norm": 3.865156874647086, | |
| "learning_rate": 6.684449266961099e-07, | |
| "loss": 0.7828, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.2641083521444696, | |
| "grad_norm": 3.298117419702452, | |
| "learning_rate": 6.451423386272311e-07, | |
| "loss": 0.792, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.309255079006772, | |
| "grad_norm": 3.5667007043819354, | |
| "learning_rate": 6.21490089951632e-07, | |
| "loss": 0.8115, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 1.3544018058690745, | |
| "grad_norm": 3.662661160157946, | |
| "learning_rate": 5.975451610080642e-07, | |
| "loss": 0.8055, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.3544018058690745, | |
| "eval_loss": 0.8635040521621704, | |
| "eval_runtime": 55.5373, | |
| "eval_samples_per_second": 56.719, | |
| "eval_steps_per_second": 0.9, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.399548532731377, | |
| "grad_norm": 3.452424328507076, | |
| "learning_rate": 5.733652372276809e-07, | |
| "loss": 0.8008, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 1.4446952595936795, | |
| "grad_norm": 3.619887615175037, | |
| "learning_rate": 5.490085701647804e-07, | |
| "loss": 0.7882, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.489841986455982, | |
| "grad_norm": 3.706405826898614, | |
| "learning_rate": 5.245338371637091e-07, | |
| "loss": 0.8086, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 1.5349887133182845, | |
| "grad_norm": 3.5950186032164213, | |
| "learning_rate": 5e-07, | |
| "loss": 0.7867, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.580135440180587, | |
| "grad_norm": 3.3585631265829803, | |
| "learning_rate": 4.75466162836291e-07, | |
| "loss": 0.7887, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.6252821670428894, | |
| "grad_norm": 3.460817031523478, | |
| "learning_rate": 4.5099142983521963e-07, | |
| "loss": 0.7808, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.670428893905192, | |
| "grad_norm": 3.400007334758639, | |
| "learning_rate": 4.2663476277231915e-07, | |
| "loss": 0.7903, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 1.7155756207674944, | |
| "grad_norm": 3.480860463387537, | |
| "learning_rate": 4.0245483899193586e-07, | |
| "loss": 0.7836, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.7607223476297968, | |
| "grad_norm": 3.589587047175883, | |
| "learning_rate": 3.785099100483681e-07, | |
| "loss": 0.7762, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 1.8058690744920993, | |
| "grad_norm": 3.6274088132542195, | |
| "learning_rate": 3.548576613727689e-07, | |
| "loss": 0.7883, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.8058690744920993, | |
| "eval_loss": 0.851189374923706, | |
| "eval_runtime": 55.4816, | |
| "eval_samples_per_second": 56.776, | |
| "eval_steps_per_second": 0.901, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.8510158013544018, | |
| "grad_norm": 3.372357535141687, | |
| "learning_rate": 3.3155507330388996e-07, | |
| "loss": 0.7971, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 1.8961625282167043, | |
| "grad_norm": 3.6104920795328432, | |
| "learning_rate": 3.086582838174551e-07, | |
| "loss": 0.7746, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.9413092550790068, | |
| "grad_norm": 3.4811562266179483, | |
| "learning_rate": 2.8622245328485907e-07, | |
| "loss": 0.7953, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 1.9864559819413092, | |
| "grad_norm": 3.3951785256342033, | |
| "learning_rate": 2.6430163158700113e-07, | |
| "loss": 0.7852, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.0316027088036117, | |
| "grad_norm": 3.7824382386544113, | |
| "learning_rate": 2.4294862790338916e-07, | |
| "loss": 0.7419, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 2.0767494356659144, | |
| "grad_norm": 3.5530924921187497, | |
| "learning_rate": 2.2221488349019902e-07, | |
| "loss": 0.7353, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.1218961625282167, | |
| "grad_norm": 3.926435566036555, | |
| "learning_rate": 2.021503477537833e-07, | |
| "loss": 0.7404, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 2.1670428893905194, | |
| "grad_norm": 3.6537792748989526, | |
| "learning_rate": 1.828033579181773e-07, | |
| "loss": 0.7461, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.2121896162528216, | |
| "grad_norm": 3.722026306821089, | |
| "learning_rate": 1.6422052257649077e-07, | |
| "loss": 0.7317, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 2.2573363431151243, | |
| "grad_norm": 3.5765809981409293, | |
| "learning_rate": 1.4644660940672627e-07, | |
| "loss": 0.7298, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.2573363431151243, | |
| "eval_loss": 0.8508689403533936, | |
| "eval_runtime": 55.5219, | |
| "eval_samples_per_second": 56.734, | |
| "eval_steps_per_second": 0.901, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.3024830699774266, | |
| "grad_norm": 3.7853317798872657, | |
| "learning_rate": 1.2952443732252054e-07, | |
| "loss": 0.7424, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 2.3476297968397293, | |
| "grad_norm": 3.7759516091418424, | |
| "learning_rate": 1.134947733186315e-07, | |
| "loss": 0.7368, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.3927765237020315, | |
| "grad_norm": 3.720889448963161, | |
| "learning_rate": 9.839623425967758e-08, | |
| "loss": 0.7319, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 2.4379232505643342, | |
| "grad_norm": 3.7006360348993064, | |
| "learning_rate": 8.426519384872732e-08, | |
| "loss": 0.7323, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.4830699774266365, | |
| "grad_norm": 3.7423771001420145, | |
| "learning_rate": 7.1135694999864e-08, | |
| "loss": 0.718, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 2.528216704288939, | |
| "grad_norm": 3.8150063518186688, | |
| "learning_rate": 5.9039367825822526e-08, | |
| "loss": 0.7243, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.5733634311512414, | |
| "grad_norm": 3.4978009765519373, | |
| "learning_rate": 4.800535343827833e-08, | |
| "loss": 0.7284, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 2.618510158013544, | |
| "grad_norm": 3.6340807089756075, | |
| "learning_rate": 3.806023374435663e-08, | |
| "loss": 0.7304, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.6636568848758464, | |
| "grad_norm": 3.8051705983031003, | |
| "learning_rate": 2.922796740848965e-08, | |
| "loss": 0.7372, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 2.708803611738149, | |
| "grad_norm": 4.590077919016769, | |
| "learning_rate": 2.1529832133895588e-08, | |
| "loss": 0.7224, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.708803611738149, | |
| "eval_loss": 0.849506676197052, | |
| "eval_runtime": 55.5608, | |
| "eval_samples_per_second": 56.695, | |
| "eval_steps_per_second": 0.9, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.7539503386004514, | |
| "grad_norm": 3.6534823714780393, | |
| "learning_rate": 1.4984373402728012e-08, | |
| "loss": 0.7413, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 2.799097065462754, | |
| "grad_norm": 3.564552616439515, | |
| "learning_rate": 9.607359798384784e-09, | |
| "loss": 0.7415, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.8442437923250563, | |
| "grad_norm": 3.9260807148979766, | |
| "learning_rate": 5.411745017609493e-09, | |
| "loss": 0.736, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 2.889390519187359, | |
| "grad_norm": 3.63819093973674, | |
| "learning_rate": 2.407636663901591e-09, | |
| "loss": 0.7229, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.9345372460496613, | |
| "grad_norm": 3.9737291373937405, | |
| "learning_rate": 6.022718974137975e-10, | |
| "loss": 0.7351, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 2.979683972911964, | |
| "grad_norm": 3.64396905938945, | |
| "learning_rate": 0.0, | |
| "loss": 0.7152, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.979683972911964, | |
| "step": 330, | |
| "total_flos": 1945595623243776.0, | |
| "train_loss": 0.8490383249340635, | |
| "train_runtime": 4686.7005, | |
| "train_samples_per_second": 18.147, | |
| "train_steps_per_second": 0.07 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 330, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 3, | |
| "save_steps": 10000, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1945595623243776.0, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |