| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 19.80456026058632, | |
| "eval_steps": 500, | |
| "global_step": 760, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.13029315960912052, | |
| "grad_norm": 0.91015625, | |
| "learning_rate": 9.998932083939657e-06, | |
| "loss": 0.7586, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.26058631921824105, | |
| "grad_norm": 0.80078125, | |
| "learning_rate": 9.995728791936505e-06, | |
| "loss": 0.7074, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.39087947882736157, | |
| "grad_norm": 0.67578125, | |
| "learning_rate": 9.990391492329341e-06, | |
| "loss": 0.6324, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.5211726384364821, | |
| "grad_norm": 0.5, | |
| "learning_rate": 9.98292246503335e-06, | |
| "loss": 0.5816, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.6514657980456026, | |
| "grad_norm": 0.361328125, | |
| "learning_rate": 9.973324900566214e-06, | |
| "loss": 0.5376, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 0.7817589576547231, | |
| "grad_norm": 0.2578125, | |
| "learning_rate": 9.961602898685225e-06, | |
| "loss": 0.5183, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.9120521172638436, | |
| "grad_norm": 0.1962890625, | |
| "learning_rate": 9.947761466636014e-06, | |
| "loss": 0.4906, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 0.990228013029316, | |
| "eval_loss": 0.4775831997394562, | |
| "eval_runtime": 216.4722, | |
| "eval_samples_per_second": 2.019, | |
| "eval_steps_per_second": 0.508, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 1.0423452768729642, | |
| "grad_norm": 0.1796875, | |
| "learning_rate": 9.931806517013612e-06, | |
| "loss": 0.4858, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 1.1726384364820848, | |
| "grad_norm": 0.1669921875, | |
| "learning_rate": 9.913744865236798e-06, | |
| "loss": 0.464, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 1.3029315960912053, | |
| "grad_norm": 0.16015625, | |
| "learning_rate": 9.893584226636773e-06, | |
| "loss": 0.4706, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.4332247557003257, | |
| "grad_norm": 0.158203125, | |
| "learning_rate": 9.871333213161438e-06, | |
| "loss": 0.4464, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 1.5635179153094463, | |
| "grad_norm": 0.1611328125, | |
| "learning_rate": 9.847001329696653e-06, | |
| "loss": 0.4411, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.6938110749185666, | |
| "grad_norm": 0.1435546875, | |
| "learning_rate": 9.820598970006068e-06, | |
| "loss": 0.4344, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 1.8241042345276872, | |
| "grad_norm": 0.1396484375, | |
| "learning_rate": 9.792137412291265e-06, | |
| "loss": 0.4292, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.9543973941368078, | |
| "grad_norm": 0.1279296875, | |
| "learning_rate": 9.761628814374074e-06, | |
| "loss": 0.3966, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 1.980456026058632, | |
| "eval_loss": 0.41222962737083435, | |
| "eval_runtime": 216.235, | |
| "eval_samples_per_second": 2.021, | |
| "eval_steps_per_second": 0.509, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 2.0846905537459284, | |
| "grad_norm": 0.126953125, | |
| "learning_rate": 9.729086208503174e-06, | |
| "loss": 0.4109, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 2.214983713355049, | |
| "grad_norm": 0.11181640625, | |
| "learning_rate": 9.694523495787149e-06, | |
| "loss": 0.3988, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 2.3452768729641695, | |
| "grad_norm": 0.1240234375, | |
| "learning_rate": 9.657955440256396e-06, | |
| "loss": 0.402, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 2.47557003257329, | |
| "grad_norm": 0.109375, | |
| "learning_rate": 9.619397662556434e-06, | |
| "loss": 0.3923, | |
| "step": 95 | |
| }, | |
| { | |
| "epoch": 2.6058631921824107, | |
| "grad_norm": 0.107421875, | |
| "learning_rate": 9.578866633275289e-06, | |
| "loss": 0.383, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 2.736156351791531, | |
| "grad_norm": 0.11083984375, | |
| "learning_rate": 9.536379665907801e-06, | |
| "loss": 0.3702, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 2.8664495114006514, | |
| "grad_norm": 0.103515625, | |
| "learning_rate": 9.491954909459895e-06, | |
| "loss": 0.3757, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.996742671009772, | |
| "grad_norm": 0.1015625, | |
| "learning_rate": 9.445611340695926e-06, | |
| "loss": 0.3648, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 2.996742671009772, | |
| "eval_loss": 0.36584609746932983, | |
| "eval_runtime": 216.4271, | |
| "eval_samples_per_second": 2.019, | |
| "eval_steps_per_second": 0.508, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 3.1270358306188926, | |
| "grad_norm": 0.09326171875, | |
| "learning_rate": 9.397368756032445e-06, | |
| "loss": 0.3651, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 3.257328990228013, | |
| "grad_norm": 0.10791015625, | |
| "learning_rate": 9.347247763081834e-06, | |
| "loss": 0.361, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 3.3876221498371337, | |
| "grad_norm": 0.08349609375, | |
| "learning_rate": 9.295269771849426e-06, | |
| "loss": 0.3513, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 3.517915309446254, | |
| "grad_norm": 0.1064453125, | |
| "learning_rate": 9.241456985587868e-06, | |
| "loss": 0.348, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 3.6482084690553744, | |
| "grad_norm": 0.083984375, | |
| "learning_rate": 9.185832391312644e-06, | |
| "loss": 0.345, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 3.778501628664495, | |
| "grad_norm": 0.076171875, | |
| "learning_rate": 9.12841974998278e-06, | |
| "loss": 0.3428, | |
| "step": 145 | |
| }, | |
| { | |
| "epoch": 3.9087947882736156, | |
| "grad_norm": 0.076171875, | |
| "learning_rate": 9.069243586350976e-06, | |
| "loss": 0.3405, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 3.986970684039088, | |
| "eval_loss": 0.33722051978111267, | |
| "eval_runtime": 216.4019, | |
| "eval_samples_per_second": 2.019, | |
| "eval_steps_per_second": 0.508, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 4.039087947882736, | |
| "grad_norm": 0.0703125, | |
| "learning_rate": 9.008329178487442e-06, | |
| "loss": 0.3302, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 4.169381107491857, | |
| "grad_norm": 0.06787109375, | |
| "learning_rate": 8.94570254698197e-06, | |
| "loss": 0.332, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 4.299674267100977, | |
| "grad_norm": 0.068359375, | |
| "learning_rate": 8.881390443828788e-06, | |
| "loss": 0.3214, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 4.429967426710098, | |
| "grad_norm": 0.06640625, | |
| "learning_rate": 8.815420340999034e-06, | |
| "loss": 0.3299, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 4.5602605863192185, | |
| "grad_norm": 0.0634765625, | |
| "learning_rate": 8.747820418705632e-06, | |
| "loss": 0.3304, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 4.690553745928339, | |
| "grad_norm": 0.08056640625, | |
| "learning_rate": 8.67861955336566e-06, | |
| "loss": 0.3249, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 4.82084690553746, | |
| "grad_norm": 0.060791015625, | |
| "learning_rate": 8.607847305265312e-06, | |
| "loss": 0.3199, | |
| "step": 185 | |
| }, | |
| { | |
| "epoch": 4.95114006514658, | |
| "grad_norm": 0.059814453125, | |
| "learning_rate": 8.535533905932739e-06, | |
| "loss": 0.3153, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 4.977198697068404, | |
| "eval_loss": 0.3180455267429352, | |
| "eval_runtime": 216.3986, | |
| "eval_samples_per_second": 2.019, | |
| "eval_steps_per_second": 0.508, | |
| "step": 191 | |
| }, | |
| { | |
| "epoch": 5.0814332247557, | |
| "grad_norm": 0.06396484375, | |
| "learning_rate": 8.461710245224149e-06, | |
| "loss": 0.3223, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 5.2117263843648205, | |
| "grad_norm": 0.060546875, | |
| "learning_rate": 8.386407858128707e-06, | |
| "loss": 0.3127, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 5.342019543973941, | |
| "grad_norm": 0.060546875, | |
| "learning_rate": 8.309658911297833e-06, | |
| "loss": 0.3092, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 5.472312703583062, | |
| "grad_norm": 0.057861328125, | |
| "learning_rate": 8.231496189304704e-06, | |
| "loss": 0.3064, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 5.602605863192182, | |
| "grad_norm": 0.05908203125, | |
| "learning_rate": 8.151953080639777e-06, | |
| "loss": 0.3119, | |
| "step": 215 | |
| }, | |
| { | |
| "epoch": 5.732899022801303, | |
| "grad_norm": 0.052734375, | |
| "learning_rate": 8.071063563448341e-06, | |
| "loss": 0.3064, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 5.863192182410423, | |
| "grad_norm": 0.0517578125, | |
| "learning_rate": 7.988862191016204e-06, | |
| "loss": 0.3035, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 5.993485342019544, | |
| "grad_norm": 0.06201171875, | |
| "learning_rate": 7.905384077009693e-06, | |
| "loss": 0.2996, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 5.993485342019544, | |
| "eval_loss": 0.30433574318885803, | |
| "eval_runtime": 216.4993, | |
| "eval_samples_per_second": 2.018, | |
| "eval_steps_per_second": 0.508, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 6.1237785016286646, | |
| "grad_norm": 0.054931640625, | |
| "learning_rate": 7.820664880476257e-06, | |
| "loss": 0.2982, | |
| "step": 235 | |
| }, | |
| { | |
| "epoch": 6.254071661237785, | |
| "grad_norm": 0.049072265625, | |
| "learning_rate": 7.734740790612137e-06, | |
| "loss": 0.3008, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 6.384364820846906, | |
| "grad_norm": 0.04931640625, | |
| "learning_rate": 7.647648511303545e-06, | |
| "loss": 0.3004, | |
| "step": 245 | |
| }, | |
| { | |
| "epoch": 6.514657980456026, | |
| "grad_norm": 0.050537109375, | |
| "learning_rate": 7.559425245448006e-06, | |
| "loss": 0.2966, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 6.644951140065147, | |
| "grad_norm": 0.05078125, | |
| "learning_rate": 7.470108679062521e-06, | |
| "loss": 0.3018, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 6.7752442996742674, | |
| "grad_norm": 0.056396484375, | |
| "learning_rate": 7.379736965185369e-06, | |
| "loss": 0.2953, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 6.905537459283387, | |
| "grad_norm": 0.04833984375, | |
| "learning_rate": 7.288348707578409e-06, | |
| "loss": 0.2987, | |
| "step": 265 | |
| }, | |
| { | |
| "epoch": 6.9837133550488595, | |
| "eval_loss": 0.2968701422214508, | |
| "eval_runtime": 216.3327, | |
| "eval_samples_per_second": 2.02, | |
| "eval_steps_per_second": 0.508, | |
| "step": 268 | |
| }, | |
| { | |
| "epoch": 7.035830618892508, | |
| "grad_norm": 0.048095703125, | |
| "learning_rate": 7.195982944236853e-06, | |
| "loss": 0.2953, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 7.166123778501628, | |
| "grad_norm": 0.05126953125, | |
| "learning_rate": 7.102679130713538e-06, | |
| "loss": 0.2936, | |
| "step": 275 | |
| }, | |
| { | |
| "epoch": 7.296416938110749, | |
| "grad_norm": 0.05322265625, | |
| "learning_rate": 7.008477123264849e-06, | |
| "loss": 0.2888, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 7.4267100977198695, | |
| "grad_norm": 0.049072265625, | |
| "learning_rate": 6.913417161825449e-06, | |
| "loss": 0.2919, | |
| "step": 285 | |
| }, | |
| { | |
| "epoch": 7.55700325732899, | |
| "grad_norm": 0.052490234375, | |
| "learning_rate": 6.817539852819149e-06, | |
| "loss": 0.2837, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 7.687296416938111, | |
| "grad_norm": 0.0458984375, | |
| "learning_rate": 6.720886151813194e-06, | |
| "loss": 0.2892, | |
| "step": 295 | |
| }, | |
| { | |
| "epoch": 7.817589576547231, | |
| "grad_norm": 0.0478515625, | |
| "learning_rate": 6.6234973460234184e-06, | |
| "loss": 0.3036, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 7.947882736156352, | |
| "grad_norm": 0.0595703125, | |
| "learning_rate": 6.525415036677745e-06, | |
| "loss": 0.2888, | |
| "step": 305 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_loss": 0.2923305630683899, | |
| "eval_runtime": 216.4767, | |
| "eval_samples_per_second": 2.019, | |
| "eval_steps_per_second": 0.508, | |
| "step": 307 | |
| }, | |
| { | |
| "epoch": 8.078175895765472, | |
| "grad_norm": 0.04248046875, | |
| "learning_rate": 6.426681121245527e-06, | |
| "loss": 0.2909, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 8.208469055374593, | |
| "grad_norm": 0.04443359375, | |
| "learning_rate": 6.327337775540362e-06, | |
| "loss": 0.2893, | |
| "step": 315 | |
| }, | |
| { | |
| "epoch": 8.338762214983714, | |
| "grad_norm": 0.042236328125, | |
| "learning_rate": 6.227427435703997e-06, | |
| "loss": 0.2981, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 8.469055374592834, | |
| "grad_norm": 0.046142578125, | |
| "learning_rate": 6.126992780079032e-06, | |
| "loss": 0.287, | |
| "step": 325 | |
| }, | |
| { | |
| "epoch": 8.599348534201955, | |
| "grad_norm": 0.042236328125, | |
| "learning_rate": 6.026076710978172e-06, | |
| "loss": 0.2904, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 8.729641693811075, | |
| "grad_norm": 0.041748046875, | |
| "learning_rate": 5.924722336357793e-06, | |
| "loss": 0.2874, | |
| "step": 335 | |
| }, | |
| { | |
| "epoch": 8.859934853420196, | |
| "grad_norm": 0.04443359375, | |
| "learning_rate": 5.82297295140367e-06, | |
| "loss": 0.2799, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 8.990228013029316, | |
| "grad_norm": 0.04345703125, | |
| "learning_rate": 5.720872020036734e-06, | |
| "loss": 0.2899, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 8.990228013029316, | |
| "eval_loss": 0.2898467779159546, | |
| "eval_runtime": 216.4655, | |
| "eval_samples_per_second": 2.019, | |
| "eval_steps_per_second": 0.508, | |
| "step": 345 | |
| }, | |
| { | |
| "epoch": 9.120521172638437, | |
| "grad_norm": 0.044189453125, | |
| "learning_rate": 5.61846315634674e-06, | |
| "loss": 0.2843, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 9.250814332247558, | |
| "grad_norm": 0.0419921875, | |
| "learning_rate": 5.515790105961785e-06, | |
| "loss": 0.2886, | |
| "step": 355 | |
| }, | |
| { | |
| "epoch": 9.381107491856678, | |
| "grad_norm": 0.044677734375, | |
| "learning_rate": 5.412896727361663e-06, | |
| "loss": 0.2898, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 9.511400651465799, | |
| "grad_norm": 0.044677734375, | |
| "learning_rate": 5.309826973142974e-06, | |
| "loss": 0.2805, | |
| "step": 365 | |
| }, | |
| { | |
| "epoch": 9.64169381107492, | |
| "grad_norm": 0.051513671875, | |
| "learning_rate": 5.206624871244066e-06, | |
| "loss": 0.2907, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 9.77198697068404, | |
| "grad_norm": 0.051513671875, | |
| "learning_rate": 5.103334506137773e-06, | |
| "loss": 0.2812, | |
| "step": 375 | |
| }, | |
| { | |
| "epoch": 9.90228013029316, | |
| "grad_norm": 0.045654296875, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2873, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 9.980456026058633, | |
| "eval_loss": 0.28830844163894653, | |
| "eval_runtime": 216.3818, | |
| "eval_samples_per_second": 2.02, | |
| "eval_steps_per_second": 0.508, | |
| "step": 383 | |
| }, | |
| { | |
| "epoch": 10.03257328990228, | |
| "grad_norm": 0.0400390625, | |
| "learning_rate": 4.89666549386223e-06, | |
| "loss": 0.2907, | |
| "step": 385 | |
| }, | |
| { | |
| "epoch": 10.1628664495114, | |
| "grad_norm": 0.04345703125, | |
| "learning_rate": 4.793375128755934e-06, | |
| "loss": 0.2848, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 10.29315960912052, | |
| "grad_norm": 0.04345703125, | |
| "learning_rate": 4.690173026857028e-06, | |
| "loss": 0.2907, | |
| "step": 395 | |
| }, | |
| { | |
| "epoch": 10.423452768729641, | |
| "grad_norm": 0.041259765625, | |
| "learning_rate": 4.587103272638339e-06, | |
| "loss": 0.2834, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 10.553745928338762, | |
| "grad_norm": 0.043701171875, | |
| "learning_rate": 4.4842098940382155e-06, | |
| "loss": 0.2862, | |
| "step": 405 | |
| }, | |
| { | |
| "epoch": 10.684039087947882, | |
| "grad_norm": 0.04638671875, | |
| "learning_rate": 4.381536843653262e-06, | |
| "loss": 0.2832, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 10.814332247557003, | |
| "grad_norm": 0.0439453125, | |
| "learning_rate": 4.279127979963266e-06, | |
| "loss": 0.2861, | |
| "step": 415 | |
| }, | |
| { | |
| "epoch": 10.944625407166123, | |
| "grad_norm": 0.040283203125, | |
| "learning_rate": 4.17702704859633e-06, | |
| "loss": 0.2831, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 10.996742671009772, | |
| "eval_loss": 0.2872494161128998, | |
| "eval_runtime": 216.4293, | |
| "eval_samples_per_second": 2.019, | |
| "eval_steps_per_second": 0.508, | |
| "step": 422 | |
| }, | |
| { | |
| "epoch": 11.074918566775244, | |
| "grad_norm": 0.03857421875, | |
| "learning_rate": 4.075277663642208e-06, | |
| "loss": 0.2913, | |
| "step": 425 | |
| }, | |
| { | |
| "epoch": 11.205211726384364, | |
| "grad_norm": 0.0439453125, | |
| "learning_rate": 3.973923289021829e-06, | |
| "loss": 0.2877, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 11.335504885993485, | |
| "grad_norm": 0.05029296875, | |
| "learning_rate": 3.8730072199209705e-06, | |
| "loss": 0.2881, | |
| "step": 435 | |
| }, | |
| { | |
| "epoch": 11.465798045602606, | |
| "grad_norm": 0.043212890625, | |
| "learning_rate": 3.7725725642960047e-06, | |
| "loss": 0.2843, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 11.596091205211726, | |
| "grad_norm": 0.0439453125, | |
| "learning_rate": 3.67266222445964e-06, | |
| "loss": 0.2822, | |
| "step": 445 | |
| }, | |
| { | |
| "epoch": 11.726384364820847, | |
| "grad_norm": 0.040771484375, | |
| "learning_rate": 3.573318878754475e-06, | |
| "loss": 0.2772, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 11.856677524429967, | |
| "grad_norm": 0.04052734375, | |
| "learning_rate": 3.4745849633222566e-06, | |
| "loss": 0.2857, | |
| "step": 455 | |
| }, | |
| { | |
| "epoch": 11.986970684039088, | |
| "grad_norm": 0.042724609375, | |
| "learning_rate": 3.3765026539765832e-06, | |
| "loss": 0.2773, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 11.986970684039088, | |
| "eval_loss": 0.286603182554245, | |
| "eval_runtime": 216.2196, | |
| "eval_samples_per_second": 2.021, | |
| "eval_steps_per_second": 0.509, | |
| "step": 460 | |
| }, | |
| { | |
| "epoch": 12.117263843648209, | |
| "grad_norm": 0.039794921875, | |
| "learning_rate": 3.2791138481868084e-06, | |
| "loss": 0.2873, | |
| "step": 465 | |
| }, | |
| { | |
| "epoch": 12.247557003257329, | |
| "grad_norm": 0.04150390625, | |
| "learning_rate": 3.1824601471808504e-06, | |
| "loss": 0.2835, | |
| "step": 470 | |
| }, | |
| { | |
| "epoch": 12.37785016286645, | |
| "grad_norm": 0.04541015625, | |
| "learning_rate": 3.0865828381745515e-06, | |
| "loss": 0.2849, | |
| "step": 475 | |
| }, | |
| { | |
| "epoch": 12.50814332247557, | |
| "grad_norm": 0.04150390625, | |
| "learning_rate": 2.991522876735154e-06, | |
| "loss": 0.2899, | |
| "step": 480 | |
| }, | |
| { | |
| "epoch": 12.63843648208469, | |
| "grad_norm": 0.040283203125, | |
| "learning_rate": 2.8973208692864623e-06, | |
| "loss": 0.2804, | |
| "step": 485 | |
| }, | |
| { | |
| "epoch": 12.768729641693811, | |
| "grad_norm": 0.03955078125, | |
| "learning_rate": 2.804017055763149e-06, | |
| "loss": 0.2792, | |
| "step": 490 | |
| }, | |
| { | |
| "epoch": 12.899022801302932, | |
| "grad_norm": 0.039794921875, | |
| "learning_rate": 2.711651292421593e-06, | |
| "loss": 0.2814, | |
| "step": 495 | |
| }, | |
| { | |
| "epoch": 12.977198697068404, | |
| "eval_loss": 0.28619444370269775, | |
| "eval_runtime": 216.4471, | |
| "eval_samples_per_second": 2.019, | |
| "eval_steps_per_second": 0.508, | |
| "step": 498 | |
| }, | |
| { | |
| "epoch": 13.029315960912053, | |
| "grad_norm": 0.046142578125, | |
| "learning_rate": 2.6202630348146323e-06, | |
| "loss": 0.283, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 13.159609120521173, | |
| "grad_norm": 0.043212890625, | |
| "learning_rate": 2.529891320937481e-06, | |
| "loss": 0.2825, | |
| "step": 505 | |
| }, | |
| { | |
| "epoch": 13.289902280130294, | |
| "grad_norm": 0.0390625, | |
| "learning_rate": 2.4405747545519966e-06, | |
| "loss": 0.2833, | |
| "step": 510 | |
| }, | |
| { | |
| "epoch": 13.420195439739414, | |
| "grad_norm": 0.038818359375, | |
| "learning_rate": 2.352351488696457e-06, | |
| "loss": 0.2798, | |
| "step": 515 | |
| }, | |
| { | |
| "epoch": 13.550488599348535, | |
| "grad_norm": 0.041259765625, | |
| "learning_rate": 2.265259209387867e-06, | |
| "loss": 0.2912, | |
| "step": 520 | |
| }, | |
| { | |
| "epoch": 13.680781758957655, | |
| "grad_norm": 0.042236328125, | |
| "learning_rate": 2.179335119523745e-06, | |
| "loss": 0.2882, | |
| "step": 525 | |
| }, | |
| { | |
| "epoch": 13.811074918566776, | |
| "grad_norm": 0.040283203125, | |
| "learning_rate": 2.094615922990309e-06, | |
| "loss": 0.2819, | |
| "step": 530 | |
| }, | |
| { | |
| "epoch": 13.941368078175895, | |
| "grad_norm": 0.040283203125, | |
| "learning_rate": 2.0111378089837958e-06, | |
| "loss": 0.2781, | |
| "step": 535 | |
| }, | |
| { | |
| "epoch": 13.993485342019543, | |
| "eval_loss": 0.2859518229961395, | |
| "eval_runtime": 216.8194, | |
| "eval_samples_per_second": 2.016, | |
| "eval_steps_per_second": 0.507, | |
| "step": 537 | |
| }, | |
| { | |
| "epoch": 14.071661237785015, | |
| "grad_norm": 0.0439453125, | |
| "learning_rate": 1.928936436551661e-06, | |
| "loss": 0.2838, | |
| "step": 540 | |
| }, | |
| { | |
| "epoch": 14.201954397394136, | |
| "grad_norm": 0.039306640625, | |
| "learning_rate": 1.848046919360225e-06, | |
| "loss": 0.2839, | |
| "step": 545 | |
| }, | |
| { | |
| "epoch": 14.332247557003257, | |
| "grad_norm": 0.041748046875, | |
| "learning_rate": 1.7685038106952952e-06, | |
| "loss": 0.279, | |
| "step": 550 | |
| }, | |
| { | |
| "epoch": 14.462540716612377, | |
| "grad_norm": 0.040771484375, | |
| "learning_rate": 1.6903410887021676e-06, | |
| "loss": 0.2772, | |
| "step": 555 | |
| }, | |
| { | |
| "epoch": 14.592833876221498, | |
| "grad_norm": 0.05224609375, | |
| "learning_rate": 1.6135921418712959e-06, | |
| "loss": 0.2817, | |
| "step": 560 | |
| }, | |
| { | |
| "epoch": 14.723127035830618, | |
| "grad_norm": 0.039306640625, | |
| "learning_rate": 1.5382897547758513e-06, | |
| "loss": 0.2932, | |
| "step": 565 | |
| }, | |
| { | |
| "epoch": 14.853420195439739, | |
| "grad_norm": 0.04150390625, | |
| "learning_rate": 1.4644660940672628e-06, | |
| "loss": 0.2834, | |
| "step": 570 | |
| }, | |
| { | |
| "epoch": 14.98371335504886, | |
| "grad_norm": 0.038330078125, | |
| "learning_rate": 1.3921526947346902e-06, | |
| "loss": 0.2845, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 14.98371335504886, | |
| "eval_loss": 0.2858189642429352, | |
| "eval_runtime": 216.4523, | |
| "eval_samples_per_second": 2.019, | |
| "eval_steps_per_second": 0.508, | |
| "step": 575 | |
| }, | |
| { | |
| "epoch": 15.11400651465798, | |
| "grad_norm": 0.040771484375, | |
| "learning_rate": 1.321380446634342e-06, | |
| "loss": 0.2778, | |
| "step": 580 | |
| }, | |
| { | |
| "epoch": 15.2442996742671, | |
| "grad_norm": 0.039306640625, | |
| "learning_rate": 1.2521795812943704e-06, | |
| "loss": 0.2812, | |
| "step": 585 | |
| }, | |
| { | |
| "epoch": 15.374592833876221, | |
| "grad_norm": 0.03759765625, | |
| "learning_rate": 1.1845796590009684e-06, | |
| "loss": 0.2834, | |
| "step": 590 | |
| }, | |
| { | |
| "epoch": 15.504885993485342, | |
| "grad_norm": 0.04150390625, | |
| "learning_rate": 1.118609556171213e-06, | |
| "loss": 0.2852, | |
| "step": 595 | |
| }, | |
| { | |
| "epoch": 15.635179153094462, | |
| "grad_norm": 0.04052734375, | |
| "learning_rate": 1.0542974530180327e-06, | |
| "loss": 0.2869, | |
| "step": 600 | |
| }, | |
| { | |
| "epoch": 15.765472312703583, | |
| "grad_norm": 0.048095703125, | |
| "learning_rate": 9.916708215125586e-07, | |
| "loss": 0.2745, | |
| "step": 605 | |
| }, | |
| { | |
| "epoch": 15.895765472312704, | |
| "grad_norm": 0.041748046875, | |
| "learning_rate": 9.307564136490255e-07, | |
| "loss": 0.29, | |
| "step": 610 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_loss": 0.28575843572616577, | |
| "eval_runtime": 216.8023, | |
| "eval_samples_per_second": 2.016, | |
| "eval_steps_per_second": 0.507, | |
| "step": 614 | |
| }, | |
| { | |
| "epoch": 16.026058631921824, | |
| "grad_norm": 0.043212890625, | |
| "learning_rate": 8.715802500172215e-07, | |
| "loss": 0.2866, | |
| "step": 615 | |
| }, | |
| { | |
| "epoch": 16.156351791530945, | |
| "grad_norm": 0.03955078125, | |
| "learning_rate": 8.141676086873574e-07, | |
| "loss": 0.2881, | |
| "step": 620 | |
| }, | |
| { | |
| "epoch": 16.286644951140065, | |
| "grad_norm": 0.048828125, | |
| "learning_rate": 7.585430144121319e-07, | |
| "loss": 0.2846, | |
| "step": 625 | |
| }, | |
| { | |
| "epoch": 16.416938110749186, | |
| "grad_norm": 0.04150390625, | |
| "learning_rate": 7.047302281505735e-07, | |
| "loss": 0.2874, | |
| "step": 630 | |
| }, | |
| { | |
| "epoch": 16.547231270358306, | |
| "grad_norm": 0.04296875, | |
| "learning_rate": 6.527522369181655e-07, | |
| "loss": 0.2818, | |
| "step": 635 | |
| }, | |
| { | |
| "epoch": 16.677524429967427, | |
| "grad_norm": 0.048828125, | |
| "learning_rate": 6.026312439675553e-07, | |
| "loss": 0.2809, | |
| "step": 640 | |
| }, | |
| { | |
| "epoch": 16.807817589576548, | |
| "grad_norm": 0.041015625, | |
| "learning_rate": 5.543886593040737e-07, | |
| "loss": 0.2831, | |
| "step": 645 | |
| }, | |
| { | |
| "epoch": 16.938110749185668, | |
| "grad_norm": 0.04052734375, | |
| "learning_rate": 5.080450905401057e-07, | |
| "loss": 0.2799, | |
| "step": 650 | |
| }, | |
| { | |
| "epoch": 16.990228013029316, | |
| "eval_loss": 0.28573134541511536, | |
| "eval_runtime": 216.4067, | |
| "eval_samples_per_second": 2.019, | |
| "eval_steps_per_second": 0.508, | |
| "step": 652 | |
| }, | |
| { | |
| "epoch": 17.06840390879479, | |
| "grad_norm": 0.0439453125, | |
| "learning_rate": 4.6362033409220077e-07, | |
| "loss": 0.2828, | |
| "step": 655 | |
| }, | |
| { | |
| "epoch": 17.19869706840391, | |
| "grad_norm": 0.047607421875, | |
| "learning_rate": 4.211333667247125e-07, | |
| "loss": 0.2811, | |
| "step": 660 | |
| }, | |
| { | |
| "epoch": 17.32899022801303, | |
| "grad_norm": 0.0478515625, | |
| "learning_rate": 3.8060233744356634e-07, | |
| "loss": 0.2881, | |
| "step": 665 | |
| }, | |
| { | |
| "epoch": 17.45928338762215, | |
| "grad_norm": 0.051025390625, | |
| "learning_rate": 3.420445597436056e-07, | |
| "loss": 0.2926, | |
| "step": 670 | |
| }, | |
| { | |
| "epoch": 17.58957654723127, | |
| "grad_norm": 0.040771484375, | |
| "learning_rate": 3.0547650421285216e-07, | |
| "loss": 0.2867, | |
| "step": 675 | |
| }, | |
| { | |
| "epoch": 17.71986970684039, | |
| "grad_norm": 0.052001953125, | |
| "learning_rate": 2.7091379149682683e-07, | |
| "loss": 0.2752, | |
| "step": 680 | |
| }, | |
| { | |
| "epoch": 17.850162866449512, | |
| "grad_norm": 0.04150390625, | |
| "learning_rate": 2.3837118562592799e-07, | |
| "loss": 0.2779, | |
| "step": 685 | |
| }, | |
| { | |
| "epoch": 17.980456026058633, | |
| "grad_norm": 0.044921875, | |
| "learning_rate": 2.0786258770873647e-07, | |
| "loss": 0.2825, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 17.980456026058633, | |
| "eval_loss": 0.28572362661361694, | |
| "eval_runtime": 216.5181, | |
| "eval_samples_per_second": 2.018, | |
| "eval_steps_per_second": 0.508, | |
| "step": 690 | |
| }, | |
| { | |
| "epoch": 18.110749185667753, | |
| "grad_norm": 0.044921875, | |
| "learning_rate": 1.7940102999393194e-07, | |
| "loss": 0.2817, | |
| "step": 695 | |
| }, | |
| { | |
| "epoch": 18.241042345276874, | |
| "grad_norm": 0.041259765625, | |
| "learning_rate": 1.5299867030334815e-07, | |
| "loss": 0.2827, | |
| "step": 700 | |
| }, | |
| { | |
| "epoch": 18.371335504885995, | |
| "grad_norm": 0.0439453125, | |
| "learning_rate": 1.286667868385627e-07, | |
| "loss": 0.2885, | |
| "step": 705 | |
| }, | |
| { | |
| "epoch": 18.501628664495115, | |
| "grad_norm": 0.05517578125, | |
| "learning_rate": 1.0641577336322761e-07, | |
| "loss": 0.2815, | |
| "step": 710 | |
| }, | |
| { | |
| "epoch": 18.631921824104236, | |
| "grad_norm": 0.03857421875, | |
| "learning_rate": 8.625513476320291e-08, | |
| "loss": 0.2849, | |
| "step": 715 | |
| }, | |
| { | |
| "epoch": 18.762214983713356, | |
| "grad_norm": 0.0419921875, | |
| "learning_rate": 6.819348298638839e-08, | |
| "loss": 0.2852, | |
| "step": 720 | |
| }, | |
| { | |
| "epoch": 18.892508143322477, | |
| "grad_norm": 0.0390625, | |
| "learning_rate": 5.223853336398632e-08, | |
| "loss": 0.2829, | |
| "step": 725 | |
| }, | |
| { | |
| "epoch": 18.996742671009773, | |
| "eval_loss": 0.28573134541511536, | |
| "eval_runtime": 216.5831, | |
| "eval_samples_per_second": 2.018, | |
| "eval_steps_per_second": 0.508, | |
| "step": 729 | |
| }, | |
| { | |
| "epoch": 19.022801302931597, | |
| "grad_norm": 0.0419921875, | |
| "learning_rate": 3.839710131477492e-08, | |
| "loss": 0.275, | |
| "step": 730 | |
| }, | |
| { | |
| "epoch": 19.153094462540718, | |
| "grad_norm": 0.04248046875, | |
| "learning_rate": 2.6675099433787212e-08, | |
| "loss": 0.2854, | |
| "step": 735 | |
| }, | |
| { | |
| "epoch": 19.28338762214984, | |
| "grad_norm": 0.04150390625, | |
| "learning_rate": 1.7077534966650767e-08, | |
| "loss": 0.2794, | |
| "step": 740 | |
| }, | |
| { | |
| "epoch": 19.41368078175896, | |
| "grad_norm": 0.03759765625, | |
| "learning_rate": 9.608507670659239e-09, | |
| "loss": 0.2782, | |
| "step": 745 | |
| }, | |
| { | |
| "epoch": 19.54397394136808, | |
| "grad_norm": 0.040283203125, | |
| "learning_rate": 4.2712080634949024e-09, | |
| "loss": 0.2878, | |
| "step": 750 | |
| }, | |
| { | |
| "epoch": 19.6742671009772, | |
| "grad_norm": 0.0439453125, | |
| "learning_rate": 1.0679160603449533e-09, | |
| "loss": 0.2894, | |
| "step": 755 | |
| }, | |
| { | |
| "epoch": 19.80456026058632, | |
| "grad_norm": 0.0439453125, | |
| "learning_rate": 0.0, | |
| "loss": 0.2801, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 19.80456026058632, | |
| "eval_loss": 0.2857275605201721, | |
| "eval_runtime": 216.4887, | |
| "eval_samples_per_second": 2.019, | |
| "eval_steps_per_second": 0.508, | |
| "step": 760 | |
| }, | |
| { | |
| "epoch": 19.80456026058632, | |
| "step": 760, | |
| "total_flos": 4.2808751411995607e+18, | |
| "train_loss": 0.32058246888612446, | |
| "train_runtime": 79494.9932, | |
| "train_samples_per_second": 0.618, | |
| "train_steps_per_second": 0.01 | |
| } | |
| ], | |
| "logging_steps": 5, | |
| "max_steps": 760, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 4.2808751411995607e+18, | |
| "train_batch_size": 4, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |