| { | |
| "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": 18.09577241357873, | |
| "learning_rate": 5e-07, | |
| "loss": 1.7252, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.09029345372460497, | |
| "grad_norm": 12.427429423392192, | |
| "learning_rate": 1e-06, | |
| "loss": 1.6001, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.13544018058690746, | |
| "grad_norm": 7.447227160720521, | |
| "learning_rate": 9.993977281025862e-07, | |
| "loss": 1.2922, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.18058690744920994, | |
| "grad_norm": 4.316398393626138, | |
| "learning_rate": 9.975923633360984e-07, | |
| "loss": 1.1376, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.22573363431151242, | |
| "grad_norm": 3.9406940968187656, | |
| "learning_rate": 9.945882549823904e-07, | |
| "loss": 1.0491, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.2708803611738149, | |
| "grad_norm": 3.810723345906196, | |
| "learning_rate": 9.90392640201615e-07, | |
| "loss": 1.0089, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.3160270880361174, | |
| "grad_norm": 3.8464121948958576, | |
| "learning_rate": 9.85015626597272e-07, | |
| "loss": 0.9962, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.3611738148984199, | |
| "grad_norm": 3.2295854401750845, | |
| "learning_rate": 9.784701678661044e-07, | |
| "loss": 0.9752, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.40632054176072235, | |
| "grad_norm": 3.8385044412155596, | |
| "learning_rate": 9.707720325915103e-07, | |
| "loss": 0.9454, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.45146726862302483, | |
| "grad_norm": 3.5887788672901273, | |
| "learning_rate": 9.619397662556433e-07, | |
| "loss": 0.9344, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.45146726862302483, | |
| "eval_loss": 0.9311009049415588, | |
| "eval_runtime": 55.6318, | |
| "eval_samples_per_second": 56.622, | |
| "eval_steps_per_second": 0.899, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.4966139954853273, | |
| "grad_norm": 3.587356821639005, | |
| "learning_rate": 9.519946465617217e-07, | |
| "loss": 0.9239, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.5417607223476298, | |
| "grad_norm": 3.32920886191508, | |
| "learning_rate": 9.409606321741774e-07, | |
| "loss": 0.9331, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.5869074492099323, | |
| "grad_norm": 3.588072493745418, | |
| "learning_rate": 9.28864305000136e-07, | |
| "loss": 0.9039, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.6320541760722348, | |
| "grad_norm": 3.503346081138309, | |
| "learning_rate": 9.157348061512726e-07, | |
| "loss": 0.8858, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.6772009029345373, | |
| "grad_norm": 3.493284507608022, | |
| "learning_rate": 9.016037657403223e-07, | |
| "loss": 0.9026, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.7223476297968398, | |
| "grad_norm": 3.3777773087899727, | |
| "learning_rate": 8.865052266813685e-07, | |
| "loss": 0.8951, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.7674943566591422, | |
| "grad_norm": 3.4055672963069994, | |
| "learning_rate": 8.704755626774795e-07, | |
| "loss": 0.8947, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.8126410835214447, | |
| "grad_norm": 3.415767069790649, | |
| "learning_rate": 8.535533905932737e-07, | |
| "loss": 0.8945, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.8577878103837472, | |
| "grad_norm": 3.625379846327293, | |
| "learning_rate": 8.357794774235092e-07, | |
| "loss": 0.8758, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.9029345372460497, | |
| "grad_norm": 3.589930432996723, | |
| "learning_rate": 8.171966420818227e-07, | |
| "loss": 0.8818, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.9029345372460497, | |
| "eval_loss": 0.8771370649337769, | |
| "eval_runtime": 55.0752, | |
| "eval_samples_per_second": 57.194, | |
| "eval_steps_per_second": 0.908, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.9480812641083521, | |
| "grad_norm": 3.431382987024524, | |
| "learning_rate": 7.978496522462167e-07, | |
| "loss": 0.8825, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.9932279909706546, | |
| "grad_norm": 3.753090530415666, | |
| "learning_rate": 7.777851165098011e-07, | |
| "loss": 0.87, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.0383747178329572, | |
| "grad_norm": 3.8762413482705185, | |
| "learning_rate": 7.570513720966107e-07, | |
| "loss": 0.8215, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 1.0835214446952597, | |
| "grad_norm": 3.6411326851223107, | |
| "learning_rate": 7.356983684129989e-07, | |
| "loss": 0.8021, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.1286681715575622, | |
| "grad_norm": 3.8015150678190293, | |
| "learning_rate": 7.13777546715141e-07, | |
| "loss": 0.8122, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 1.1738148984198646, | |
| "grad_norm": 3.554114852631137, | |
| "learning_rate": 6.913417161825449e-07, | |
| "loss": 0.812, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 1.2189616252821671, | |
| "grad_norm": 3.512124531880888, | |
| "learning_rate": 6.684449266961099e-07, | |
| "loss": 0.7826, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 1.2641083521444696, | |
| "grad_norm": 3.6396827015577697, | |
| "learning_rate": 6.451423386272311e-07, | |
| "loss": 0.7936, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 1.309255079006772, | |
| "grad_norm": 3.6873566290287325, | |
| "learning_rate": 6.21490089951632e-07, | |
| "loss": 0.8014, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 1.3544018058690745, | |
| "grad_norm": 3.6757170742910867, | |
| "learning_rate": 5.975451610080642e-07, | |
| "loss": 0.7974, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.3544018058690745, | |
| "eval_loss": 0.8601916432380676, | |
| "eval_runtime": 54.9398, | |
| "eval_samples_per_second": 57.335, | |
| "eval_steps_per_second": 0.91, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 1.399548532731377, | |
| "grad_norm": 3.4471162121206023, | |
| "learning_rate": 5.733652372276809e-07, | |
| "loss": 0.7972, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 1.4446952595936795, | |
| "grad_norm": 3.545405581828385, | |
| "learning_rate": 5.490085701647804e-07, | |
| "loss": 0.8086, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 1.489841986455982, | |
| "grad_norm": 3.500167727908525, | |
| "learning_rate": 5.245338371637091e-07, | |
| "loss": 0.7844, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 1.5349887133182845, | |
| "grad_norm": 3.683324105059045, | |
| "learning_rate": 5e-07, | |
| "loss": 0.814, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 1.580135440180587, | |
| "grad_norm": 3.5076105573904224, | |
| "learning_rate": 4.75466162836291e-07, | |
| "loss": 0.7606, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 1.6252821670428894, | |
| "grad_norm": 3.39732342028104, | |
| "learning_rate": 4.5099142983521963e-07, | |
| "loss": 0.7836, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 1.670428893905192, | |
| "grad_norm": 3.522550690645414, | |
| "learning_rate": 4.2663476277231915e-07, | |
| "loss": 0.7882, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 1.7155756207674944, | |
| "grad_norm": 3.896694774670664, | |
| "learning_rate": 4.0245483899193586e-07, | |
| "loss": 0.8049, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 1.7607223476297968, | |
| "grad_norm": 3.9732787046924556, | |
| "learning_rate": 3.785099100483681e-07, | |
| "loss": 0.7794, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 1.8058690744920993, | |
| "grad_norm": 3.8350374340781825, | |
| "learning_rate": 3.548576613727689e-07, | |
| "loss": 0.7942, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.8058690744920993, | |
| "eval_loss": 0.8489097356796265, | |
| "eval_runtime": 55.1124, | |
| "eval_samples_per_second": 57.156, | |
| "eval_steps_per_second": 0.907, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 1.8510158013544018, | |
| "grad_norm": 3.647701362803135, | |
| "learning_rate": 3.3155507330388996e-07, | |
| "loss": 0.7827, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 1.8961625282167043, | |
| "grad_norm": 3.5156707572378187, | |
| "learning_rate": 3.086582838174551e-07, | |
| "loss": 0.7899, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 1.9413092550790068, | |
| "grad_norm": 3.56378125431085, | |
| "learning_rate": 2.8622245328485907e-07, | |
| "loss": 0.8011, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 1.9864559819413092, | |
| "grad_norm": 3.465941533238677, | |
| "learning_rate": 2.6430163158700113e-07, | |
| "loss": 0.7722, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 2.0316027088036117, | |
| "grad_norm": 3.780773444000903, | |
| "learning_rate": 2.4294862790338916e-07, | |
| "loss": 0.7563, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 2.0767494356659144, | |
| "grad_norm": 3.58554801379656, | |
| "learning_rate": 2.2221488349019902e-07, | |
| "loss": 0.7356, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 2.1218961625282167, | |
| "grad_norm": 3.7489661478667906, | |
| "learning_rate": 2.021503477537833e-07, | |
| "loss": 0.7422, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 2.1670428893905194, | |
| "grad_norm": 3.6327819309767717, | |
| "learning_rate": 1.828033579181773e-07, | |
| "loss": 0.7331, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 2.2121896162528216, | |
| "grad_norm": 3.862265193047436, | |
| "learning_rate": 1.6422052257649077e-07, | |
| "loss": 0.7378, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 2.2573363431151243, | |
| "grad_norm": 3.7313891351245787, | |
| "learning_rate": 1.4644660940672627e-07, | |
| "loss": 0.7405, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.2573363431151243, | |
| "eval_loss": 0.8493446111679077, | |
| "eval_runtime": 54.9903, | |
| "eval_samples_per_second": 57.283, | |
| "eval_steps_per_second": 0.909, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 2.3024830699774266, | |
| "grad_norm": 3.6025441770788005, | |
| "learning_rate": 1.2952443732252054e-07, | |
| "loss": 0.7311, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 2.3476297968397293, | |
| "grad_norm": 3.7220315278082143, | |
| "learning_rate": 1.134947733186315e-07, | |
| "loss": 0.7183, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 2.3927765237020315, | |
| "grad_norm": 4.011429432621815, | |
| "learning_rate": 9.839623425967758e-08, | |
| "loss": 0.7381, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 2.4379232505643342, | |
| "grad_norm": 3.5985031594527155, | |
| "learning_rate": 8.426519384872732e-08, | |
| "loss": 0.7359, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 2.4830699774266365, | |
| "grad_norm": 3.8934477742892417, | |
| "learning_rate": 7.1135694999864e-08, | |
| "loss": 0.7423, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 2.528216704288939, | |
| "grad_norm": 4.140166060538144, | |
| "learning_rate": 5.9039367825822526e-08, | |
| "loss": 0.7336, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 2.5733634311512414, | |
| "grad_norm": 3.685670529512933, | |
| "learning_rate": 4.800535343827833e-08, | |
| "loss": 0.7231, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 2.618510158013544, | |
| "grad_norm": 3.764984332523118, | |
| "learning_rate": 3.806023374435663e-08, | |
| "loss": 0.731, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 2.6636568848758464, | |
| "grad_norm": 3.695489561322323, | |
| "learning_rate": 2.922796740848965e-08, | |
| "loss": 0.7243, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 2.708803611738149, | |
| "grad_norm": 3.785952827138228, | |
| "learning_rate": 2.1529832133895588e-08, | |
| "loss": 0.7321, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.708803611738149, | |
| "eval_loss": 0.8476512432098389, | |
| "eval_runtime": 55.0992, | |
| "eval_samples_per_second": 57.17, | |
| "eval_steps_per_second": 0.907, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 2.7539503386004514, | |
| "grad_norm": 3.719298487431947, | |
| "learning_rate": 1.4984373402728012e-08, | |
| "loss": 0.7248, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 2.799097065462754, | |
| "grad_norm": 3.570380389588657, | |
| "learning_rate": 9.607359798384784e-09, | |
| "loss": 0.7527, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 2.8442437923250563, | |
| "grad_norm": 3.547493528277377, | |
| "learning_rate": 5.411745017609493e-09, | |
| "loss": 0.738, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 2.889390519187359, | |
| "grad_norm": 3.7382434901187755, | |
| "learning_rate": 2.407636663901591e-09, | |
| "loss": 0.7268, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 2.9345372460496613, | |
| "grad_norm": 3.5368152248539033, | |
| "learning_rate": 6.022718974137975e-10, | |
| "loss": 0.7383, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 2.979683972911964, | |
| "grad_norm": 3.5267604120974374, | |
| "learning_rate": 0.0, | |
| "loss": 0.7248, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 2.979683972911964, | |
| "step": 330, | |
| "total_flos": 1945595623243776.0, | |
| "train_loss": 0.8492823528520989, | |
| "train_runtime": 4699.828, | |
| "train_samples_per_second": 18.096, | |
| "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 | |
| } | |