ss-en-m2m100-combo / trainer_state.json
richardlastrucci's picture
added model files
03411af
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 3.0,
"eval_steps": 500,
"global_step": 39924,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.04,
"learning_rate": 4.9373810239454966e-05,
"loss": 3.0783,
"step": 500
},
{
"epoch": 0.08,
"learning_rate": 4.874762047890993e-05,
"loss": 2.5238,
"step": 1000
},
{
"epoch": 0.11,
"learning_rate": 4.8121430718364894e-05,
"loss": 2.2994,
"step": 1500
},
{
"epoch": 0.15,
"learning_rate": 4.749524095781986e-05,
"loss": 2.1655,
"step": 2000
},
{
"epoch": 0.19,
"learning_rate": 4.686905119727483e-05,
"loss": 2.1007,
"step": 2500
},
{
"epoch": 0.23,
"learning_rate": 4.624286143672979e-05,
"loss": 2.0447,
"step": 3000
},
{
"epoch": 0.26,
"learning_rate": 4.5616671676184756e-05,
"loss": 1.9793,
"step": 3500
},
{
"epoch": 0.3,
"learning_rate": 4.499048191563971e-05,
"loss": 1.928,
"step": 4000
},
{
"epoch": 0.34,
"learning_rate": 4.436429215509468e-05,
"loss": 1.8916,
"step": 4500
},
{
"epoch": 0.38,
"learning_rate": 4.373810239454965e-05,
"loss": 1.8331,
"step": 5000
},
{
"epoch": 0.41,
"learning_rate": 4.311191263400461e-05,
"loss": 1.8014,
"step": 5500
},
{
"epoch": 0.45,
"learning_rate": 4.2485722873459575e-05,
"loss": 1.7924,
"step": 6000
},
{
"epoch": 0.49,
"learning_rate": 4.185953311291454e-05,
"loss": 1.7729,
"step": 6500
},
{
"epoch": 0.53,
"learning_rate": 4.12333433523695e-05,
"loss": 1.7446,
"step": 7000
},
{
"epoch": 0.56,
"learning_rate": 4.060715359182447e-05,
"loss": 1.7415,
"step": 7500
},
{
"epoch": 0.6,
"learning_rate": 3.9980963831279437e-05,
"loss": 1.7042,
"step": 8000
},
{
"epoch": 0.64,
"learning_rate": 3.9354774070734394e-05,
"loss": 1.678,
"step": 8500
},
{
"epoch": 0.68,
"learning_rate": 3.8728584310189364e-05,
"loss": 1.6576,
"step": 9000
},
{
"epoch": 0.71,
"learning_rate": 3.810239454964433e-05,
"loss": 1.6765,
"step": 9500
},
{
"epoch": 0.75,
"learning_rate": 3.747620478909929e-05,
"loss": 1.6109,
"step": 10000
},
{
"epoch": 0.79,
"learning_rate": 3.6850015028554255e-05,
"loss": 1.6197,
"step": 10500
},
{
"epoch": 0.83,
"learning_rate": 3.622382526800922e-05,
"loss": 1.6011,
"step": 11000
},
{
"epoch": 0.86,
"learning_rate": 3.559763550746418e-05,
"loss": 1.5907,
"step": 11500
},
{
"epoch": 0.9,
"learning_rate": 3.497144574691915e-05,
"loss": 1.5875,
"step": 12000
},
{
"epoch": 0.94,
"learning_rate": 3.434525598637412e-05,
"loss": 1.5752,
"step": 12500
},
{
"epoch": 0.98,
"learning_rate": 3.3719066225829074e-05,
"loss": 1.5237,
"step": 13000
},
{
"epoch": 1.01,
"learning_rate": 3.309287646528404e-05,
"loss": 1.4724,
"step": 13500
},
{
"epoch": 1.05,
"learning_rate": 3.246668670473901e-05,
"loss": 1.3053,
"step": 14000
},
{
"epoch": 1.09,
"learning_rate": 3.184049694419397e-05,
"loss": 1.3191,
"step": 14500
},
{
"epoch": 1.13,
"learning_rate": 3.1214307183648936e-05,
"loss": 1.2972,
"step": 15000
},
{
"epoch": 1.16,
"learning_rate": 3.05881174231039e-05,
"loss": 1.2705,
"step": 15500
},
{
"epoch": 1.2,
"learning_rate": 2.9961927662558864e-05,
"loss": 1.3129,
"step": 16000
},
{
"epoch": 1.24,
"learning_rate": 2.9335737902013828e-05,
"loss": 1.3122,
"step": 16500
},
{
"epoch": 1.28,
"learning_rate": 2.8709548141468788e-05,
"loss": 1.2842,
"step": 17000
},
{
"epoch": 1.31,
"learning_rate": 2.808335838092376e-05,
"loss": 1.2766,
"step": 17500
},
{
"epoch": 1.35,
"learning_rate": 2.745716862037872e-05,
"loss": 1.2869,
"step": 18000
},
{
"epoch": 1.39,
"learning_rate": 2.6830978859833683e-05,
"loss": 1.2822,
"step": 18500
},
{
"epoch": 1.43,
"learning_rate": 2.620478909928865e-05,
"loss": 1.2513,
"step": 19000
},
{
"epoch": 1.47,
"learning_rate": 2.5578599338743614e-05,
"loss": 1.2716,
"step": 19500
},
{
"epoch": 1.5,
"learning_rate": 2.495240957819858e-05,
"loss": 1.2569,
"step": 20000
},
{
"epoch": 1.54,
"learning_rate": 2.432621981765354e-05,
"loss": 1.2537,
"step": 20500
},
{
"epoch": 1.58,
"learning_rate": 2.370003005710851e-05,
"loss": 1.26,
"step": 21000
},
{
"epoch": 1.62,
"learning_rate": 2.307384029656347e-05,
"loss": 1.2559,
"step": 21500
},
{
"epoch": 1.65,
"learning_rate": 2.2447650536018436e-05,
"loss": 1.2259,
"step": 22000
},
{
"epoch": 1.69,
"learning_rate": 2.18214607754734e-05,
"loss": 1.2254,
"step": 22500
},
{
"epoch": 1.73,
"learning_rate": 2.1195271014928364e-05,
"loss": 1.2075,
"step": 23000
},
{
"epoch": 1.77,
"learning_rate": 2.056908125438333e-05,
"loss": 1.2135,
"step": 23500
},
{
"epoch": 1.8,
"learning_rate": 1.9942891493838295e-05,
"loss": 1.2082,
"step": 24000
},
{
"epoch": 1.84,
"learning_rate": 1.931670173329326e-05,
"loss": 1.1962,
"step": 24500
},
{
"epoch": 1.88,
"learning_rate": 1.8690511972748222e-05,
"loss": 1.2139,
"step": 25000
},
{
"epoch": 1.92,
"learning_rate": 1.8064322212203186e-05,
"loss": 1.1852,
"step": 25500
},
{
"epoch": 1.95,
"learning_rate": 1.743813245165815e-05,
"loss": 1.1937,
"step": 26000
},
{
"epoch": 1.99,
"learning_rate": 1.6811942691113117e-05,
"loss": 1.1962,
"step": 26500
},
{
"epoch": 2.03,
"learning_rate": 1.618575293056808e-05,
"loss": 1.0053,
"step": 27000
},
{
"epoch": 2.07,
"learning_rate": 1.5559563170023044e-05,
"loss": 0.9642,
"step": 27500
},
{
"epoch": 2.1,
"learning_rate": 1.4933373409478008e-05,
"loss": 0.9592,
"step": 28000
},
{
"epoch": 2.14,
"learning_rate": 1.4307183648932974e-05,
"loss": 0.9827,
"step": 28500
},
{
"epoch": 2.18,
"learning_rate": 1.368099388838794e-05,
"loss": 0.9665,
"step": 29000
},
{
"epoch": 2.22,
"learning_rate": 1.3054804127842901e-05,
"loss": 0.9826,
"step": 29500
},
{
"epoch": 2.25,
"learning_rate": 1.2428614367297867e-05,
"loss": 0.9679,
"step": 30000
},
{
"epoch": 2.29,
"learning_rate": 1.180242460675283e-05,
"loss": 0.9684,
"step": 30500
},
{
"epoch": 2.33,
"learning_rate": 1.1176234846207794e-05,
"loss": 0.9709,
"step": 31000
},
{
"epoch": 2.37,
"learning_rate": 1.055004508566276e-05,
"loss": 0.9489,
"step": 31500
},
{
"epoch": 2.4,
"learning_rate": 9.923855325117724e-06,
"loss": 0.9516,
"step": 32000
},
{
"epoch": 2.44,
"learning_rate": 9.297665564572689e-06,
"loss": 0.9691,
"step": 32500
},
{
"epoch": 2.48,
"learning_rate": 8.671475804027653e-06,
"loss": 0.9417,
"step": 33000
},
{
"epoch": 2.52,
"learning_rate": 8.045286043482618e-06,
"loss": 0.9505,
"step": 33500
},
{
"epoch": 2.55,
"learning_rate": 7.419096282937582e-06,
"loss": 0.9594,
"step": 34000
},
{
"epoch": 2.59,
"learning_rate": 6.792906522392546e-06,
"loss": 0.9334,
"step": 34500
},
{
"epoch": 2.63,
"learning_rate": 6.1667167618475106e-06,
"loss": 0.9486,
"step": 35000
},
{
"epoch": 2.67,
"learning_rate": 5.540527001302475e-06,
"loss": 0.9476,
"step": 35500
},
{
"epoch": 2.71,
"learning_rate": 4.914337240757439e-06,
"loss": 0.9366,
"step": 36000
},
{
"epoch": 2.74,
"learning_rate": 4.288147480212404e-06,
"loss": 0.9253,
"step": 36500
},
{
"epoch": 2.78,
"learning_rate": 3.661957719667368e-06,
"loss": 0.9279,
"step": 37000
},
{
"epoch": 2.82,
"learning_rate": 3.035767959122333e-06,
"loss": 0.9246,
"step": 37500
},
{
"epoch": 2.86,
"learning_rate": 2.409578198577297e-06,
"loss": 0.923,
"step": 38000
},
{
"epoch": 2.89,
"learning_rate": 1.7833884380322613e-06,
"loss": 0.9241,
"step": 38500
},
{
"epoch": 2.93,
"learning_rate": 1.157198677487226e-06,
"loss": 0.9359,
"step": 39000
},
{
"epoch": 2.97,
"learning_rate": 5.310089169421901e-07,
"loss": 0.9117,
"step": 39500
},
{
"epoch": 3.0,
"step": 39924,
"total_flos": 5.321321329491149e+16,
"train_loss": 1.3535074337774975,
"train_runtime": 29786.9631,
"train_samples_per_second": 10.722,
"train_steps_per_second": 1.34
}
],
"logging_steps": 500,
"max_steps": 39924,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 10000,
"total_flos": 5.321321329491149e+16,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}