| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 0.14450867052023122, | |
| "eval_steps": 100, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.000361271676300578, | |
| "eval_loss": 1.744890570640564, | |
| "eval_runtime": 95.2541, | |
| "eval_samples_per_second": 12.241, | |
| "eval_steps_per_second": 6.12, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0018063583815028901, | |
| "grad_norm": 4.56119966506958, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 6.469, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0036127167630057803, | |
| "grad_norm": 5.107720375061035, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 5.8106, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00541907514450867, | |
| "grad_norm": 6.6674933433532715, | |
| "learning_rate": 5e-05, | |
| "loss": 6.2328, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0072254335260115606, | |
| "grad_norm": 5.628376007080078, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 6.5034, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.009031791907514452, | |
| "grad_norm": 6.5742106437683105, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 6.1949, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01083815028901734, | |
| "grad_norm": 7.4489922523498535, | |
| "learning_rate": 0.0001, | |
| "loss": 6.8727, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.012644508670520232, | |
| "grad_norm": 7.205574989318848, | |
| "learning_rate": 9.995494831023409e-05, | |
| "loss": 5.8499, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.014450867052023121, | |
| "grad_norm": 10.276925086975098, | |
| "learning_rate": 9.981987442712633e-05, | |
| "loss": 6.6391, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.01625722543352601, | |
| "grad_norm": 10.676612854003906, | |
| "learning_rate": 9.959502176294383e-05, | |
| "loss": 6.1377, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.018063583815028903, | |
| "grad_norm": 21.650407791137695, | |
| "learning_rate": 9.928079551738543e-05, | |
| "loss": 6.4974, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.019869942196531792, | |
| "grad_norm": 4.22016716003418, | |
| "learning_rate": 9.887776194738432e-05, | |
| "loss": 5.6711, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.02167630057803468, | |
| "grad_norm": 3.700309991836548, | |
| "learning_rate": 9.838664734667495e-05, | |
| "loss": 5.6352, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.02348265895953757, | |
| "grad_norm": 4.459054946899414, | |
| "learning_rate": 9.780833673696254e-05, | |
| "loss": 5.9549, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.025289017341040464, | |
| "grad_norm": 5.048811912536621, | |
| "learning_rate": 9.714387227305422e-05, | |
| "loss": 6.3563, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.027095375722543353, | |
| "grad_norm": 5.600136756896973, | |
| "learning_rate": 9.639445136482548e-05, | |
| "loss": 6.4251, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.028901734104046242, | |
| "grad_norm": 7.410740375518799, | |
| "learning_rate": 9.55614245194068e-05, | |
| "loss": 5.9274, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.03070809248554913, | |
| "grad_norm": 6.403213024139404, | |
| "learning_rate": 9.464629290747842e-05, | |
| "loss": 6.1228, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.03251445086705202, | |
| "grad_norm": 7.519406318664551, | |
| "learning_rate": 9.365070565805941e-05, | |
| "loss": 6.6063, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.034320809248554913, | |
| "grad_norm": 9.153055191040039, | |
| "learning_rate": 9.257645688666556e-05, | |
| "loss": 6.4336, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.036127167630057806, | |
| "grad_norm": 23.724929809570312, | |
| "learning_rate": 9.142548246219212e-05, | |
| "loss": 6.556, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.036127167630057806, | |
| "eval_loss": 1.5442149639129639, | |
| "eval_runtime": 96.4401, | |
| "eval_samples_per_second": 12.09, | |
| "eval_steps_per_second": 6.045, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.03793352601156069, | |
| "grad_norm": 4.142440319061279, | |
| "learning_rate": 9.019985651834703e-05, | |
| "loss": 5.4367, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.039739884393063585, | |
| "grad_norm": 3.686530351638794, | |
| "learning_rate": 8.890178771592199e-05, | |
| "loss": 5.9418, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.04154624277456647, | |
| "grad_norm": 3.7836296558380127, | |
| "learning_rate": 8.753361526263621e-05, | |
| "loss": 5.8358, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.04335260115606936, | |
| "grad_norm": 4.974203586578369, | |
| "learning_rate": 8.609780469772623e-05, | |
| "loss": 6.6595, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.045158959537572256, | |
| "grad_norm": 6.760814189910889, | |
| "learning_rate": 8.459694344887732e-05, | |
| "loss": 5.4935, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.04696531791907514, | |
| "grad_norm": 5.244045734405518, | |
| "learning_rate": 8.303373616950408e-05, | |
| "loss": 5.8446, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.048771676300578035, | |
| "grad_norm": 7.215941905975342, | |
| "learning_rate": 8.141099986478212e-05, | |
| "loss": 6.7863, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.05057803468208093, | |
| "grad_norm": 6.362617492675781, | |
| "learning_rate": 7.973165881521434e-05, | |
| "loss": 6.3179, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.05238439306358381, | |
| "grad_norm": 8.617600440979004, | |
| "learning_rate": 7.799873930687978e-05, | |
| "loss": 5.7684, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.054190751445086706, | |
| "grad_norm": 15.83110523223877, | |
| "learning_rate": 7.621536417786159e-05, | |
| "loss": 6.5653, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.0559971098265896, | |
| "grad_norm": 3.476327896118164, | |
| "learning_rate": 7.438474719068173e-05, | |
| "loss": 5.8152, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.057803468208092484, | |
| "grad_norm": 3.5889389514923096, | |
| "learning_rate": 7.251018724088367e-05, | |
| "loss": 5.7597, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.05960982658959538, | |
| "grad_norm": 4.485110759735107, | |
| "learning_rate": 7.059506241219965e-05, | |
| "loss": 5.8891, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.06141618497109826, | |
| "grad_norm": 5.032641887664795, | |
| "learning_rate": 6.864282388901544e-05, | |
| "loss": 5.7698, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.06322254335260116, | |
| "grad_norm": 6.4402594566345215, | |
| "learning_rate": 6.665698973710288e-05, | |
| "loss": 5.5398, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.06502890173410404, | |
| "grad_norm": 5.370179653167725, | |
| "learning_rate": 6.464113856382752e-05, | |
| "loss": 6.0105, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.06683526011560693, | |
| "grad_norm": 5.48384428024292, | |
| "learning_rate": 6.259890306925627e-05, | |
| "loss": 5.4653, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.06864161849710983, | |
| "grad_norm": 7.3742594718933105, | |
| "learning_rate": 6.0533963499786314e-05, | |
| "loss": 5.9431, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.07044797687861272, | |
| "grad_norm": 10.979525566101074, | |
| "learning_rate": 5.8450041016092464e-05, | |
| "loss": 6.3899, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.07225433526011561, | |
| "grad_norm": 18.570680618286133, | |
| "learning_rate": 5.6350890987343944e-05, | |
| "loss": 7.3082, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.07225433526011561, | |
| "eval_loss": 1.511340618133545, | |
| "eval_runtime": 96.1463, | |
| "eval_samples_per_second": 12.127, | |
| "eval_steps_per_second": 6.064, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.07406069364161849, | |
| "grad_norm": 3.3652424812316895, | |
| "learning_rate": 5.4240296223775465e-05, | |
| "loss": 5.5137, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.07586705202312138, | |
| "grad_norm": 3.8494818210601807, | |
| "learning_rate": 5.212206015980742e-05, | |
| "loss": 6.0826, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.07767341040462428, | |
| "grad_norm": 3.5938096046447754, | |
| "learning_rate": 5e-05, | |
| "loss": 6.0547, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.07947976878612717, | |
| "grad_norm": 5.043278694152832, | |
| "learning_rate": 4.78779398401926e-05, | |
| "loss": 5.8113, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.08128612716763006, | |
| "grad_norm": 5.376712799072266, | |
| "learning_rate": 4.575970377622456e-05, | |
| "loss": 5.8342, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.08309248554913294, | |
| "grad_norm": 6.200654029846191, | |
| "learning_rate": 4.364910901265606e-05, | |
| "loss": 5.86, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.08489884393063583, | |
| "grad_norm": 7.157764434814453, | |
| "learning_rate": 4.1549958983907555e-05, | |
| "loss": 5.8018, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.08670520231213873, | |
| "grad_norm": 8.071253776550293, | |
| "learning_rate": 3.94660365002137e-05, | |
| "loss": 6.3807, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.08851156069364162, | |
| "grad_norm": 7.438620567321777, | |
| "learning_rate": 3.740109693074375e-05, | |
| "loss": 5.9327, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.09031791907514451, | |
| "grad_norm": 14.404324531555176, | |
| "learning_rate": 3.5358861436172485e-05, | |
| "loss": 6.5125, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.0921242774566474, | |
| "grad_norm": 2.9339845180511475, | |
| "learning_rate": 3.334301026289712e-05, | |
| "loss": 5.8859, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.09393063583815028, | |
| "grad_norm": 3.615297317504883, | |
| "learning_rate": 3.135717611098458e-05, | |
| "loss": 5.6365, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.09573699421965318, | |
| "grad_norm": 3.7128686904907227, | |
| "learning_rate": 2.9404937587800375e-05, | |
| "loss": 5.6875, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.09754335260115607, | |
| "grad_norm": 4.845981597900391, | |
| "learning_rate": 2.748981275911633e-05, | |
| "loss": 5.1678, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.09934971098265896, | |
| "grad_norm": 5.1541008949279785, | |
| "learning_rate": 2.5615252809318284e-05, | |
| "loss": 5.5239, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.10115606936416185, | |
| "grad_norm": 5.836907863616943, | |
| "learning_rate": 2.3784635822138424e-05, | |
| "loss": 5.7149, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.10296242774566473, | |
| "grad_norm": 7.401642322540283, | |
| "learning_rate": 2.2001260693120233e-05, | |
| "loss": 6.4186, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.10476878612716763, | |
| "grad_norm": 8.060465812683105, | |
| "learning_rate": 2.026834118478567e-05, | |
| "loss": 5.5685, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.10657514450867052, | |
| "grad_norm": 8.320117950439453, | |
| "learning_rate": 1.858900013521788e-05, | |
| "loss": 5.8179, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.10838150289017341, | |
| "grad_norm": 18.933137893676758, | |
| "learning_rate": 1.6966263830495936e-05, | |
| "loss": 6.4365, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.10838150289017341, | |
| "eval_loss": 1.500579595565796, | |
| "eval_runtime": 96.4129, | |
| "eval_samples_per_second": 12.094, | |
| "eval_steps_per_second": 6.047, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.1101878612716763, | |
| "grad_norm": 3.131831407546997, | |
| "learning_rate": 1.5403056551122697e-05, | |
| "loss": 5.1589, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.1119942196531792, | |
| "grad_norm": 4.786662578582764, | |
| "learning_rate": 1.3902195302273779e-05, | |
| "loss": 5.8762, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.11380057803468208, | |
| "grad_norm": 3.8581812381744385, | |
| "learning_rate": 1.246638473736378e-05, | |
| "loss": 6.0422, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.11560693641618497, | |
| "grad_norm": 5.307023525238037, | |
| "learning_rate": 1.1098212284078036e-05, | |
| "loss": 6.0744, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.11741329479768786, | |
| "grad_norm": 4.369593143463135, | |
| "learning_rate": 9.800143481652979e-06, | |
| "loss": 6.31, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.11921965317919075, | |
| "grad_norm": 5.19207763671875, | |
| "learning_rate": 8.574517537807897e-06, | |
| "loss": 5.7647, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.12102601156069365, | |
| "grad_norm": 5.733678340911865, | |
| "learning_rate": 7.423543113334436e-06, | |
| "loss": 6.1191, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.12283236994219653, | |
| "grad_norm": 6.88820219039917, | |
| "learning_rate": 6.349294341940593e-06, | |
| "loss": 6.7561, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.12463872832369942, | |
| "grad_norm": 7.364393711090088, | |
| "learning_rate": 5.353707092521582e-06, | |
| "loss": 5.9415, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 0.12644508670520233, | |
| "grad_norm": 21.01556968688965, | |
| "learning_rate": 4.43857548059321e-06, | |
| "loss": 5.9318, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.1282514450867052, | |
| "grad_norm": 3.3074729442596436, | |
| "learning_rate": 3.605548635174533e-06, | |
| "loss": 5.5233, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 0.13005780346820808, | |
| "grad_norm": 3.396397829055786, | |
| "learning_rate": 2.85612772694579e-06, | |
| "loss": 5.9158, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.13186416184971098, | |
| "grad_norm": 3.6206588745117188, | |
| "learning_rate": 2.191663263037458e-06, | |
| "loss": 5.5862, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 0.13367052023121387, | |
| "grad_norm": 4.75822114944458, | |
| "learning_rate": 1.6133526533250565e-06, | |
| "loss": 6.0164, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.13547687861271676, | |
| "grad_norm": 6.198365688323975, | |
| "learning_rate": 1.1222380526156928e-06, | |
| "loss": 5.7811, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.13728323699421965, | |
| "grad_norm": 6.1432647705078125, | |
| "learning_rate": 7.192044826145771e-07, | |
| "loss": 5.7833, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.13908959537572255, | |
| "grad_norm": 5.783768177032471, | |
| "learning_rate": 4.049782370561583e-07, | |
| "loss": 6.0198, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 0.14089595375722544, | |
| "grad_norm": 5.420467853546143, | |
| "learning_rate": 1.8012557287367392e-07, | |
| "loss": 5.5835, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.14270231213872833, | |
| "grad_norm": 7.798616409301758, | |
| "learning_rate": 4.5051689765929214e-08, | |
| "loss": 6.2971, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 0.14450867052023122, | |
| "grad_norm": 12.888833999633789, | |
| "learning_rate": 0.0, | |
| "loss": 6.3966, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.14450867052023122, | |
| "eval_loss": 1.4953255653381348, | |
| "eval_runtime": 96.3977, | |
| "eval_samples_per_second": 12.096, | |
| "eval_steps_per_second": 6.048, | |
| "step": 400 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 400, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 1, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 1.4301801881586893e+17, | |
| "train_batch_size": 2, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |