| { | |
| "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": 95.6948, | |
| "eval_samples_per_second": 12.185, | |
| "eval_steps_per_second": 6.092, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.0018063583815028901, | |
| "grad_norm": 4.680508613586426, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 6.4694, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.0036127167630057803, | |
| "grad_norm": 5.228878498077393, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 5.8076, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.00541907514450867, | |
| "grad_norm": 6.814301490783691, | |
| "learning_rate": 5e-05, | |
| "loss": 6.2301, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.0072254335260115606, | |
| "grad_norm": 5.734597206115723, | |
| "learning_rate": 6.666666666666667e-05, | |
| "loss": 6.5014, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.009031791907514452, | |
| "grad_norm": 6.919921875, | |
| "learning_rate": 8.333333333333334e-05, | |
| "loss": 6.1907, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.01083815028901734, | |
| "grad_norm": 7.7186737060546875, | |
| "learning_rate": 0.0001, | |
| "loss": 6.8673, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.012644508670520232, | |
| "grad_norm": 7.3959550857543945, | |
| "learning_rate": 9.995494831023409e-05, | |
| "loss": 5.8464, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.014450867052023121, | |
| "grad_norm": 10.59113597869873, | |
| "learning_rate": 9.981987442712633e-05, | |
| "loss": 6.6395, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.01625722543352601, | |
| "grad_norm": 10.594133377075195, | |
| "learning_rate": 9.959502176294383e-05, | |
| "loss": 6.1328, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 0.018063583815028903, | |
| "grad_norm": 22.168792724609375, | |
| "learning_rate": 9.928079551738543e-05, | |
| "loss": 6.4779, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.019869942196531792, | |
| "grad_norm": 4.262362480163574, | |
| "learning_rate": 9.887776194738432e-05, | |
| "loss": 5.6685, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 0.02167630057803468, | |
| "grad_norm": 3.6753509044647217, | |
| "learning_rate": 9.838664734667495e-05, | |
| "loss": 5.6305, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 0.02348265895953757, | |
| "grad_norm": 4.582398891448975, | |
| "learning_rate": 9.780833673696254e-05, | |
| "loss": 5.9505, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 0.025289017341040464, | |
| "grad_norm": 5.06224250793457, | |
| "learning_rate": 9.714387227305422e-05, | |
| "loss": 6.3539, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 0.027095375722543353, | |
| "grad_norm": 5.636335372924805, | |
| "learning_rate": 9.639445136482548e-05, | |
| "loss": 6.4256, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 0.028901734104046242, | |
| "grad_norm": 7.639087200164795, | |
| "learning_rate": 9.55614245194068e-05, | |
| "loss": 5.9327, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 0.03070809248554913, | |
| "grad_norm": 6.552780628204346, | |
| "learning_rate": 9.464629290747842e-05, | |
| "loss": 6.1245, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 0.03251445086705202, | |
| "grad_norm": 7.416606426239014, | |
| "learning_rate": 9.365070565805941e-05, | |
| "loss": 6.6008, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 0.034320809248554913, | |
| "grad_norm": 9.532649040222168, | |
| "learning_rate": 9.257645688666556e-05, | |
| "loss": 6.4374, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 0.036127167630057806, | |
| "grad_norm": 19.954425811767578, | |
| "learning_rate": 9.142548246219212e-05, | |
| "loss": 6.5905, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.036127167630057806, | |
| "eval_loss": 1.5467851161956787, | |
| "eval_runtime": 97.1043, | |
| "eval_samples_per_second": 12.008, | |
| "eval_steps_per_second": 6.004, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 0.03793352601156069, | |
| "grad_norm": 4.287236213684082, | |
| "learning_rate": 9.019985651834703e-05, | |
| "loss": 5.4495, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 0.039739884393063585, | |
| "grad_norm": 3.7286770343780518, | |
| "learning_rate": 8.890178771592199e-05, | |
| "loss": 5.9449, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 0.04154624277456647, | |
| "grad_norm": 3.869879722595215, | |
| "learning_rate": 8.753361526263621e-05, | |
| "loss": 5.8386, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 0.04335260115606936, | |
| "grad_norm": 4.93843936920166, | |
| "learning_rate": 8.609780469772623e-05, | |
| "loss": 6.6573, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 0.045158959537572256, | |
| "grad_norm": 6.8780131340026855, | |
| "learning_rate": 8.459694344887732e-05, | |
| "loss": 5.4958, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 0.04696531791907514, | |
| "grad_norm": 5.2665629386901855, | |
| "learning_rate": 8.303373616950408e-05, | |
| "loss": 5.8536, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 0.048771676300578035, | |
| "grad_norm": 7.371923446655273, | |
| "learning_rate": 8.141099986478212e-05, | |
| "loss": 6.7899, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 0.05057803468208093, | |
| "grad_norm": 6.423787593841553, | |
| "learning_rate": 7.973165881521434e-05, | |
| "loss": 6.321, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 0.05238439306358381, | |
| "grad_norm": 8.676210403442383, | |
| "learning_rate": 7.799873930687978e-05, | |
| "loss": 5.7517, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 0.054190751445086706, | |
| "grad_norm": 16.351991653442383, | |
| "learning_rate": 7.621536417786159e-05, | |
| "loss": 6.5635, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 0.0559971098265896, | |
| "grad_norm": 6.630072116851807, | |
| "learning_rate": 7.438474719068173e-05, | |
| "loss": 5.8176, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 0.057803468208092484, | |
| "grad_norm": 3.5932905673980713, | |
| "learning_rate": 7.251018724088367e-05, | |
| "loss": 5.7641, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 0.05960982658959538, | |
| "grad_norm": 4.526276588439941, | |
| "learning_rate": 7.059506241219965e-05, | |
| "loss": 5.89, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 0.06141618497109826, | |
| "grad_norm": 5.0801897048950195, | |
| "learning_rate": 6.864282388901544e-05, | |
| "loss": 5.7756, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 0.06322254335260116, | |
| "grad_norm": 6.393936634063721, | |
| "learning_rate": 6.665698973710288e-05, | |
| "loss": 5.5407, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 0.06502890173410404, | |
| "grad_norm": 5.460756778717041, | |
| "learning_rate": 6.464113856382752e-05, | |
| "loss": 6.0142, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 0.06683526011560693, | |
| "grad_norm": 5.452981472015381, | |
| "learning_rate": 6.259890306925627e-05, | |
| "loss": 5.4672, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 0.06864161849710983, | |
| "grad_norm": 7.4432549476623535, | |
| "learning_rate": 6.0533963499786314e-05, | |
| "loss": 5.9396, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 0.07044797687861272, | |
| "grad_norm": 11.204219818115234, | |
| "learning_rate": 5.8450041016092464e-05, | |
| "loss": 6.3922, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 0.07225433526011561, | |
| "grad_norm": 21.239145278930664, | |
| "learning_rate": 5.6350890987343944e-05, | |
| "loss": 7.3291, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.07225433526011561, | |
| "eval_loss": 1.5103161334991455, | |
| "eval_runtime": 96.7823, | |
| "eval_samples_per_second": 12.048, | |
| "eval_steps_per_second": 6.024, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 0.07406069364161849, | |
| "grad_norm": 3.3259732723236084, | |
| "learning_rate": 5.4240296223775465e-05, | |
| "loss": 5.5065, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 0.07586705202312138, | |
| "grad_norm": 3.890929937362671, | |
| "learning_rate": 5.212206015980742e-05, | |
| "loss": 6.083, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 0.07767341040462428, | |
| "grad_norm": 3.593712091445923, | |
| "learning_rate": 5e-05, | |
| "loss": 6.0546, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 0.07947976878612717, | |
| "grad_norm": 5.011513710021973, | |
| "learning_rate": 4.78779398401926e-05, | |
| "loss": 5.8085, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 0.08128612716763006, | |
| "grad_norm": 5.653621196746826, | |
| "learning_rate": 4.575970377622456e-05, | |
| "loss": 5.8344, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 0.08309248554913294, | |
| "grad_norm": 6.257894992828369, | |
| "learning_rate": 4.364910901265606e-05, | |
| "loss": 5.8473, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 0.08489884393063583, | |
| "grad_norm": 7.058216571807861, | |
| "learning_rate": 4.1549958983907555e-05, | |
| "loss": 5.7983, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 0.08670520231213873, | |
| "grad_norm": 7.803954601287842, | |
| "learning_rate": 3.94660365002137e-05, | |
| "loss": 6.376, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 0.08851156069364162, | |
| "grad_norm": 7.779985427856445, | |
| "learning_rate": 3.740109693074375e-05, | |
| "loss": 5.9278, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 0.09031791907514451, | |
| "grad_norm": 14.273958206176758, | |
| "learning_rate": 3.5358861436172485e-05, | |
| "loss": 6.5011, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 0.0921242774566474, | |
| "grad_norm": 2.9535694122314453, | |
| "learning_rate": 3.334301026289712e-05, | |
| "loss": 5.8836, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 0.09393063583815028, | |
| "grad_norm": 3.5993616580963135, | |
| "learning_rate": 3.135717611098458e-05, | |
| "loss": 5.6311, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 0.09573699421965318, | |
| "grad_norm": 3.7176356315612793, | |
| "learning_rate": 2.9404937587800375e-05, | |
| "loss": 5.6883, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 0.09754335260115607, | |
| "grad_norm": 4.789780139923096, | |
| "learning_rate": 2.748981275911633e-05, | |
| "loss": 5.1705, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 0.09934971098265896, | |
| "grad_norm": 5.1265082359313965, | |
| "learning_rate": 2.5615252809318284e-05, | |
| "loss": 5.5257, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 0.10115606936416185, | |
| "grad_norm": 5.957849025726318, | |
| "learning_rate": 2.3784635822138424e-05, | |
| "loss": 5.7171, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 0.10296242774566473, | |
| "grad_norm": 7.608188629150391, | |
| "learning_rate": 2.2001260693120233e-05, | |
| "loss": 6.4109, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 0.10476878612716763, | |
| "grad_norm": 8.30679988861084, | |
| "learning_rate": 2.026834118478567e-05, | |
| "loss": 5.5608, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 0.10657514450867052, | |
| "grad_norm": 7.934511184692383, | |
| "learning_rate": 1.858900013521788e-05, | |
| "loss": 5.8088, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 0.10838150289017341, | |
| "grad_norm": 21.605234146118164, | |
| "learning_rate": 1.6966263830495936e-05, | |
| "loss": 6.426, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.10838150289017341, | |
| "eval_loss": 1.499989628791809, | |
| "eval_runtime": 96.913, | |
| "eval_samples_per_second": 12.031, | |
| "eval_steps_per_second": 6.016, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 0.1101878612716763, | |
| "grad_norm": 3.113095760345459, | |
| "learning_rate": 1.5403056551122697e-05, | |
| "loss": 5.1541, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 0.1119942196531792, | |
| "grad_norm": 3.644343852996826, | |
| "learning_rate": 1.3902195302273779e-05, | |
| "loss": 5.8757, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 0.11380057803468208, | |
| "grad_norm": 3.9557442665100098, | |
| "learning_rate": 1.246638473736378e-05, | |
| "loss": 6.0388, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 0.11560693641618497, | |
| "grad_norm": 5.263237476348877, | |
| "learning_rate": 1.1098212284078036e-05, | |
| "loss": 6.0708, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 0.11741329479768786, | |
| "grad_norm": 4.272531986236572, | |
| "learning_rate": 9.800143481652979e-06, | |
| "loss": 6.3018, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 0.11921965317919075, | |
| "grad_norm": 5.232715129852295, | |
| "learning_rate": 8.574517537807897e-06, | |
| "loss": 5.764, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 0.12102601156069365, | |
| "grad_norm": 5.70760440826416, | |
| "learning_rate": 7.423543113334436e-06, | |
| "loss": 6.1271, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 0.12283236994219653, | |
| "grad_norm": 6.9662184715271, | |
| "learning_rate": 6.349294341940593e-06, | |
| "loss": 6.7533, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 0.12463872832369942, | |
| "grad_norm": 7.42333984375, | |
| "learning_rate": 5.353707092521582e-06, | |
| "loss": 5.9473, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 0.12644508670520233, | |
| "grad_norm": 21.6599178314209, | |
| "learning_rate": 4.43857548059321e-06, | |
| "loss": 5.9251, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 0.1282514450867052, | |
| "grad_norm": 3.270684003829956, | |
| "learning_rate": 3.605548635174533e-06, | |
| "loss": 5.5248, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 0.13005780346820808, | |
| "grad_norm": 3.4227538108825684, | |
| "learning_rate": 2.85612772694579e-06, | |
| "loss": 5.9192, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 0.13186416184971098, | |
| "grad_norm": 3.6004385948181152, | |
| "learning_rate": 2.191663263037458e-06, | |
| "loss": 5.5829, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 0.13367052023121387, | |
| "grad_norm": 4.801909923553467, | |
| "learning_rate": 1.6133526533250565e-06, | |
| "loss": 6.0279, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 0.13547687861271676, | |
| "grad_norm": 6.258386611938477, | |
| "learning_rate": 1.1222380526156928e-06, | |
| "loss": 5.7802, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 0.13728323699421965, | |
| "grad_norm": 6.687994956970215, | |
| "learning_rate": 7.192044826145771e-07, | |
| "loss": 5.7802, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 0.13908959537572255, | |
| "grad_norm": 5.807911396026611, | |
| "learning_rate": 4.049782370561583e-07, | |
| "loss": 6.0196, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 0.14089595375722544, | |
| "grad_norm": 5.457380294799805, | |
| "learning_rate": 1.8012557287367392e-07, | |
| "loss": 5.5855, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 0.14270231213872833, | |
| "grad_norm": 7.9476118087768555, | |
| "learning_rate": 4.5051689765929214e-08, | |
| "loss": 6.2977, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 0.14450867052023122, | |
| "grad_norm": 12.78293228149414, | |
| "learning_rate": 0.0, | |
| "loss": 6.387, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 0.14450867052023122, | |
| "eval_loss": 1.4946010112762451, | |
| "eval_runtime": 97.1081, | |
| "eval_samples_per_second": 12.007, | |
| "eval_steps_per_second": 6.004, | |
| "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 | |
| } | |