open-llama-chat-3k / trainer_state.json
Andreas Köpf
init model
1b46b37
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.8107741059302853,
"global_step": 3000,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.01,
"learning_rate": 4.5e-08,
"loss": 1.8279,
"step": 10
},
{
"epoch": 0.01,
"learning_rate": 9.5e-08,
"loss": 1.7023,
"step": 20
},
{
"epoch": 0.02,
"learning_rate": 1.4500000000000001e-07,
"loss": 1.6779,
"step": 30
},
{
"epoch": 0.02,
"learning_rate": 1.95e-07,
"loss": 1.6776,
"step": 40
},
{
"epoch": 0.03,
"learning_rate": 2.4500000000000004e-07,
"loss": 1.6379,
"step": 50
},
{
"epoch": 0.04,
"learning_rate": 2.9500000000000003e-07,
"loss": 1.5523,
"step": 60
},
{
"epoch": 0.04,
"learning_rate": 3.4500000000000003e-07,
"loss": 1.3693,
"step": 70
},
{
"epoch": 0.05,
"learning_rate": 3.9500000000000003e-07,
"loss": 1.3299,
"step": 80
},
{
"epoch": 0.05,
"learning_rate": 4.4500000000000003e-07,
"loss": 1.2452,
"step": 90
},
{
"epoch": 0.06,
"learning_rate": 4.95e-07,
"loss": 1.25,
"step": 100
},
{
"epoch": 0.07,
"learning_rate": 5.450000000000001e-07,
"loss": 1.1376,
"step": 110
},
{
"epoch": 0.07,
"learning_rate": 5.95e-07,
"loss": 1.1166,
"step": 120
},
{
"epoch": 0.08,
"learning_rate": 6.450000000000001e-07,
"loss": 1.1069,
"step": 130
},
{
"epoch": 0.08,
"learning_rate": 6.950000000000001e-07,
"loss": 1.0921,
"step": 140
},
{
"epoch": 0.09,
"learning_rate": 7.450000000000001e-07,
"loss": 1.0674,
"step": 150
},
{
"epoch": 0.1,
"learning_rate": 7.950000000000001e-07,
"loss": 1.0283,
"step": 160
},
{
"epoch": 0.1,
"learning_rate": 8.450000000000002e-07,
"loss": 1.0252,
"step": 170
},
{
"epoch": 0.11,
"learning_rate": 8.95e-07,
"loss": 0.993,
"step": 180
},
{
"epoch": 0.11,
"learning_rate": 9.450000000000001e-07,
"loss": 0.9808,
"step": 190
},
{
"epoch": 0.12,
"learning_rate": 9.950000000000002e-07,
"loss": 0.9879,
"step": 200
},
{
"epoch": 0.13,
"learning_rate": 1.045e-06,
"loss": 0.9724,
"step": 210
},
{
"epoch": 0.13,
"learning_rate": 1.095e-06,
"loss": 0.9544,
"step": 220
},
{
"epoch": 0.14,
"learning_rate": 1.145e-06,
"loss": 0.9349,
"step": 230
},
{
"epoch": 0.14,
"learning_rate": 1.195e-06,
"loss": 0.9247,
"step": 240
},
{
"epoch": 0.15,
"learning_rate": 1.2450000000000002e-06,
"loss": 0.9251,
"step": 250
},
{
"epoch": 0.16,
"learning_rate": 1.295e-06,
"loss": 0.938,
"step": 260
},
{
"epoch": 0.16,
"learning_rate": 1.3450000000000003e-06,
"loss": 0.8967,
"step": 270
},
{
"epoch": 0.17,
"learning_rate": 1.3950000000000002e-06,
"loss": 0.9132,
"step": 280
},
{
"epoch": 0.18,
"learning_rate": 1.445e-06,
"loss": 0.9081,
"step": 290
},
{
"epoch": 0.18,
"learning_rate": 1.495e-06,
"loss": 0.8873,
"step": 300
},
{
"epoch": 0.19,
"learning_rate": 1.545e-06,
"loss": 0.8978,
"step": 310
},
{
"epoch": 0.19,
"learning_rate": 1.5950000000000002e-06,
"loss": 0.8628,
"step": 320
},
{
"epoch": 0.2,
"learning_rate": 1.6450000000000001e-06,
"loss": 0.8651,
"step": 330
},
{
"epoch": 0.21,
"learning_rate": 1.6950000000000003e-06,
"loss": 0.8499,
"step": 340
},
{
"epoch": 0.21,
"learning_rate": 1.745e-06,
"loss": 0.8514,
"step": 350
},
{
"epoch": 0.22,
"learning_rate": 1.7950000000000002e-06,
"loss": 0.8645,
"step": 360
},
{
"epoch": 0.22,
"learning_rate": 1.8450000000000001e-06,
"loss": 0.8632,
"step": 370
},
{
"epoch": 0.23,
"learning_rate": 1.895e-06,
"loss": 0.8382,
"step": 380
},
{
"epoch": 0.24,
"learning_rate": 1.945e-06,
"loss": 0.8531,
"step": 390
},
{
"epoch": 0.24,
"learning_rate": 1.9950000000000004e-06,
"loss": 0.8413,
"step": 400
},
{
"epoch": 0.25,
"learning_rate": 2.045e-06,
"loss": 0.8272,
"step": 410
},
{
"epoch": 0.25,
"learning_rate": 2.0950000000000003e-06,
"loss": 0.831,
"step": 420
},
{
"epoch": 0.26,
"learning_rate": 2.1450000000000002e-06,
"loss": 0.8077,
"step": 430
},
{
"epoch": 0.27,
"learning_rate": 2.195e-06,
"loss": 0.8209,
"step": 440
},
{
"epoch": 0.27,
"learning_rate": 2.245e-06,
"loss": 0.8249,
"step": 450
},
{
"epoch": 0.28,
"learning_rate": 2.2950000000000005e-06,
"loss": 0.8037,
"step": 460
},
{
"epoch": 0.28,
"learning_rate": 2.345e-06,
"loss": 0.7976,
"step": 470
},
{
"epoch": 0.29,
"learning_rate": 2.395e-06,
"loss": 0.8046,
"step": 480
},
{
"epoch": 0.3,
"learning_rate": 2.4450000000000003e-06,
"loss": 0.7959,
"step": 490
},
{
"epoch": 0.3,
"learning_rate": 2.4950000000000003e-06,
"loss": 0.8176,
"step": 500
},
{
"epoch": 0.3,
"eval_oasst_export_accuracy": 0.7042471551142071,
"eval_oasst_export_loss": 1.2229081392288208,
"eval_oasst_export_runtime": 68.987,
"eval_oasst_export_samples_per_second": 30.354,
"eval_oasst_export_steps_per_second": 1.276,
"step": 500
},
{
"epoch": 0.3,
"eval_code_alpaca_accuracy": 0.8437448762092147,
"eval_code_alpaca_loss": 0.5453789234161377,
"eval_code_alpaca_runtime": 4.9365,
"eval_code_alpaca_samples_per_second": 50.643,
"eval_code_alpaca_steps_per_second": 2.228,
"step": 500
},
{
"epoch": 0.3,
"eval_evol_v2_accuracy": 0.7902228855879827,
"eval_evol_v2_loss": 0.7061598300933838,
"eval_evol_v2_runtime": 275.1097,
"eval_evol_v2_samples_per_second": 25.99,
"eval_evol_v2_steps_per_second": 1.083,
"step": 500
},
{
"epoch": 0.3,
"eval_nlu_instruct_accuracy": 0.7843999034325105,
"eval_nlu_instruct_loss": 0.757986843585968,
"eval_nlu_instruct_runtime": 3477.7847,
"eval_nlu_instruct_samples_per_second": 22.435,
"eval_nlu_instruct_steps_per_second": 0.935,
"step": 500
},
{
"epoch": 0.3,
"eval_grade_school_math_instructions_accuracy": 0.7889496290511518,
"eval_grade_school_math_instructions_loss": 0.7259232997894287,
"eval_grade_school_math_instructions_runtime": 10.7804,
"eval_grade_school_math_instructions_samples_per_second": 40.815,
"eval_grade_school_math_instructions_steps_per_second": 1.762,
"step": 500
},
{
"epoch": 0.3,
"eval_poem_instructions_accuracy": 0.49874821391497626,
"eval_poem_instructions_loss": 2.590395450592041,
"eval_poem_instructions_runtime": 21.1843,
"eval_poem_instructions_samples_per_second": 16.38,
"eval_poem_instructions_steps_per_second": 0.708,
"step": 500
},
{
"epoch": 0.3,
"eval_gpt4all_accuracy": 0.7821919414875455,
"eval_gpt4all_loss": 0.7756699323654175,
"eval_gpt4all_runtime": 3565.0056,
"eval_gpt4all_samples_per_second": 21.813,
"eval_gpt4all_steps_per_second": 0.909,
"step": 500
},
{
"epoch": 0.3,
"eval_joke_accuracy": 0.5321712147485625,
"eval_joke_loss": 1.970703125,
"eval_joke_runtime": 5.8841,
"eval_joke_samples_per_second": 12.916,
"eval_joke_steps_per_second": 0.68,
"step": 500
},
{
"epoch": 0.3,
"eval_gsm8k_accuracy": 0.8398790847611236,
"eval_gsm8k_loss": 0.5526099801063538,
"eval_gsm8k_runtime": 22.7523,
"eval_gsm8k_samples_per_second": 57.972,
"eval_gsm8k_steps_per_second": 2.417,
"step": 500
},
{
"epoch": 0.31,
"learning_rate": 2.545e-06,
"loss": 0.8266,
"step": 510
},
{
"epoch": 0.31,
"learning_rate": 2.595e-06,
"loss": 0.8252,
"step": 520
},
{
"epoch": 0.32,
"learning_rate": 2.6450000000000005e-06,
"loss": 0.806,
"step": 530
},
{
"epoch": 0.33,
"learning_rate": 2.6950000000000005e-06,
"loss": 0.7994,
"step": 540
},
{
"epoch": 0.33,
"learning_rate": 2.7450000000000004e-06,
"loss": 0.8062,
"step": 550
},
{
"epoch": 0.34,
"learning_rate": 2.7950000000000003e-06,
"loss": 0.8086,
"step": 560
},
{
"epoch": 0.34,
"learning_rate": 2.845e-06,
"loss": 0.8055,
"step": 570
},
{
"epoch": 0.35,
"learning_rate": 2.8950000000000002e-06,
"loss": 0.7825,
"step": 580
},
{
"epoch": 0.36,
"learning_rate": 2.945e-06,
"loss": 0.796,
"step": 590
},
{
"epoch": 0.36,
"learning_rate": 2.995e-06,
"loss": 0.7974,
"step": 600
},
{
"epoch": 0.37,
"learning_rate": 3.045e-06,
"loss": 0.8071,
"step": 610
},
{
"epoch": 0.37,
"learning_rate": 3.0950000000000004e-06,
"loss": 0.7967,
"step": 620
},
{
"epoch": 0.38,
"learning_rate": 3.1450000000000004e-06,
"loss": 0.778,
"step": 630
},
{
"epoch": 0.39,
"learning_rate": 3.1950000000000003e-06,
"loss": 0.7869,
"step": 640
},
{
"epoch": 0.39,
"learning_rate": 3.2450000000000003e-06,
"loss": 0.7828,
"step": 650
},
{
"epoch": 0.4,
"learning_rate": 3.2950000000000002e-06,
"loss": 0.7927,
"step": 660
},
{
"epoch": 0.4,
"learning_rate": 3.3450000000000006e-06,
"loss": 0.7709,
"step": 670
},
{
"epoch": 0.41,
"learning_rate": 3.3950000000000005e-06,
"loss": 0.7624,
"step": 680
},
{
"epoch": 0.42,
"learning_rate": 3.445e-06,
"loss": 0.7739,
"step": 690
},
{
"epoch": 0.42,
"learning_rate": 3.495e-06,
"loss": 0.77,
"step": 700
},
{
"epoch": 0.43,
"learning_rate": 3.545e-06,
"loss": 0.7814,
"step": 710
},
{
"epoch": 0.43,
"learning_rate": 3.5950000000000003e-06,
"loss": 0.7822,
"step": 720
},
{
"epoch": 0.44,
"learning_rate": 3.6450000000000003e-06,
"loss": 0.7816,
"step": 730
},
{
"epoch": 0.45,
"learning_rate": 3.695e-06,
"loss": 0.7935,
"step": 740
},
{
"epoch": 0.45,
"learning_rate": 3.745e-06,
"loss": 0.7738,
"step": 750
},
{
"epoch": 0.46,
"learning_rate": 3.7950000000000005e-06,
"loss": 0.7585,
"step": 760
},
{
"epoch": 0.46,
"learning_rate": 3.8450000000000005e-06,
"loss": 0.7776,
"step": 770
},
{
"epoch": 0.47,
"learning_rate": 3.895000000000001e-06,
"loss": 0.7681,
"step": 780
},
{
"epoch": 0.48,
"learning_rate": 3.945e-06,
"loss": 0.7836,
"step": 790
},
{
"epoch": 0.48,
"learning_rate": 3.995000000000001e-06,
"loss": 0.7687,
"step": 800
},
{
"epoch": 0.49,
"learning_rate": 4.045e-06,
"loss": 0.7852,
"step": 810
},
{
"epoch": 0.49,
"learning_rate": 4.095e-06,
"loss": 0.753,
"step": 820
},
{
"epoch": 0.5,
"learning_rate": 4.145e-06,
"loss": 0.7676,
"step": 830
},
{
"epoch": 0.51,
"learning_rate": 4.1950000000000005e-06,
"loss": 0.7626,
"step": 840
},
{
"epoch": 0.51,
"learning_rate": 4.245e-06,
"loss": 0.7578,
"step": 850
},
{
"epoch": 0.52,
"learning_rate": 4.295e-06,
"loss": 0.7743,
"step": 860
},
{
"epoch": 0.53,
"learning_rate": 4.345000000000001e-06,
"loss": 0.7662,
"step": 870
},
{
"epoch": 0.53,
"learning_rate": 4.395e-06,
"loss": 0.7709,
"step": 880
},
{
"epoch": 0.54,
"learning_rate": 4.445000000000001e-06,
"loss": 0.7638,
"step": 890
},
{
"epoch": 0.54,
"learning_rate": 4.495e-06,
"loss": 0.752,
"step": 900
},
{
"epoch": 0.55,
"learning_rate": 4.5450000000000005e-06,
"loss": 0.7744,
"step": 910
},
{
"epoch": 0.56,
"learning_rate": 4.595000000000001e-06,
"loss": 0.7607,
"step": 920
},
{
"epoch": 0.56,
"learning_rate": 4.645e-06,
"loss": 0.765,
"step": 930
},
{
"epoch": 0.57,
"learning_rate": 4.695e-06,
"loss": 0.7534,
"step": 940
},
{
"epoch": 0.57,
"learning_rate": 4.745e-06,
"loss": 0.7569,
"step": 950
},
{
"epoch": 0.58,
"learning_rate": 4.795e-06,
"loss": 0.7522,
"step": 960
},
{
"epoch": 0.59,
"learning_rate": 4.845e-06,
"loss": 0.7876,
"step": 970
},
{
"epoch": 0.59,
"learning_rate": 4.8950000000000006e-06,
"loss": 0.7586,
"step": 980
},
{
"epoch": 0.6,
"learning_rate": 4.945e-06,
"loss": 0.7658,
"step": 990
},
{
"epoch": 0.6,
"learning_rate": 4.9950000000000005e-06,
"loss": 0.7562,
"step": 1000
},
{
"epoch": 0.6,
"eval_oasst_export_accuracy": 0.7149679693032294,
"eval_oasst_export_loss": 1.1956202983856201,
"eval_oasst_export_runtime": 67.0497,
"eval_oasst_export_samples_per_second": 31.231,
"eval_oasst_export_steps_per_second": 1.312,
"step": 1000
},
{
"epoch": 0.6,
"eval_code_alpaca_accuracy": 0.8463136033229491,
"eval_code_alpaca_loss": 0.5398945212364197,
"eval_code_alpaca_runtime": 6.22,
"eval_code_alpaca_samples_per_second": 40.193,
"eval_code_alpaca_steps_per_second": 1.769,
"step": 1000
},
{
"epoch": 0.6,
"eval_evol_v2_accuracy": 0.7982416515833893,
"eval_evol_v2_loss": 0.6806697845458984,
"eval_evol_v2_runtime": 274.9092,
"eval_evol_v2_samples_per_second": 26.009,
"eval_evol_v2_steps_per_second": 1.084,
"step": 1000
},
{
"epoch": 0.6,
"eval_nlu_instruct_accuracy": 0.8084633890177162,
"eval_nlu_instruct_loss": 0.6906304955482483,
"eval_nlu_instruct_runtime": 3476.832,
"eval_nlu_instruct_samples_per_second": 22.441,
"eval_nlu_instruct_steps_per_second": 0.935,
"step": 1000
},
{
"epoch": 0.6,
"eval_grade_school_math_instructions_accuracy": 0.8071065989847716,
"eval_grade_school_math_instructions_loss": 0.6586337089538574,
"eval_grade_school_math_instructions_runtime": 9.647,
"eval_grade_school_math_instructions_samples_per_second": 45.61,
"eval_grade_school_math_instructions_steps_per_second": 1.97,
"step": 1000
},
{
"epoch": 0.6,
"eval_poem_instructions_accuracy": 0.506542535644907,
"eval_poem_instructions_loss": 2.5567362308502197,
"eval_poem_instructions_runtime": 21.1754,
"eval_poem_instructions_samples_per_second": 16.387,
"eval_poem_instructions_steps_per_second": 0.708,
"step": 1000
},
{
"epoch": 0.6,
"eval_gpt4all_accuracy": 0.7917010462623065,
"eval_gpt4all_loss": 0.7404432892799377,
"eval_gpt4all_runtime": 3565.553,
"eval_gpt4all_samples_per_second": 21.809,
"eval_gpt4all_steps_per_second": 0.909,
"step": 1000
},
{
"epoch": 0.6,
"eval_joke_accuracy": 0.5601898329834809,
"eval_joke_loss": 1.8026316165924072,
"eval_joke_runtime": 4.4938,
"eval_joke_samples_per_second": 16.912,
"eval_joke_steps_per_second": 0.89,
"step": 1000
},
{
"epoch": 0.6,
"eval_gsm8k_accuracy": 0.8582343392395189,
"eval_gsm8k_loss": 0.4936060607433319,
"eval_gsm8k_runtime": 22.7991,
"eval_gsm8k_samples_per_second": 57.853,
"eval_gsm8k_steps_per_second": 2.412,
"step": 1000
},
{
"epoch": 0.61,
"learning_rate": 5.045e-06,
"loss": 0.7579,
"step": 1010
},
{
"epoch": 0.62,
"learning_rate": 5.095e-06,
"loss": 0.7684,
"step": 1020
},
{
"epoch": 0.62,
"learning_rate": 5.145e-06,
"loss": 0.7464,
"step": 1030
},
{
"epoch": 0.63,
"learning_rate": 5.195e-06,
"loss": 0.7691,
"step": 1040
},
{
"epoch": 0.63,
"learning_rate": 5.245e-06,
"loss": 0.7385,
"step": 1050
},
{
"epoch": 0.64,
"learning_rate": 5.295e-06,
"loss": 0.7704,
"step": 1060
},
{
"epoch": 0.65,
"learning_rate": 5.3450000000000005e-06,
"loss": 0.7735,
"step": 1070
},
{
"epoch": 0.65,
"learning_rate": 5.395e-06,
"loss": 0.7448,
"step": 1080
},
{
"epoch": 0.66,
"learning_rate": 5.445e-06,
"loss": 0.7358,
"step": 1090
},
{
"epoch": 0.66,
"learning_rate": 5.495000000000001e-06,
"loss": 0.762,
"step": 1100
},
{
"epoch": 0.67,
"learning_rate": 5.545e-06,
"loss": 0.7636,
"step": 1110
},
{
"epoch": 0.68,
"learning_rate": 5.595000000000001e-06,
"loss": 0.7662,
"step": 1120
},
{
"epoch": 0.68,
"learning_rate": 5.645e-06,
"loss": 0.7565,
"step": 1130
},
{
"epoch": 0.69,
"learning_rate": 5.6950000000000005e-06,
"loss": 0.7564,
"step": 1140
},
{
"epoch": 0.69,
"learning_rate": 5.745000000000001e-06,
"loss": 0.7576,
"step": 1150
},
{
"epoch": 0.7,
"learning_rate": 5.795e-06,
"loss": 0.7588,
"step": 1160
},
{
"epoch": 0.71,
"learning_rate": 5.845000000000001e-06,
"loss": 0.7503,
"step": 1170
},
{
"epoch": 0.71,
"learning_rate": 5.895e-06,
"loss": 0.7579,
"step": 1180
},
{
"epoch": 0.72,
"learning_rate": 5.945000000000001e-06,
"loss": 0.772,
"step": 1190
},
{
"epoch": 0.72,
"learning_rate": 5.995000000000001e-06,
"loss": 0.7507,
"step": 1200
},
{
"epoch": 0.73,
"learning_rate": 6.0450000000000006e-06,
"loss": 0.7448,
"step": 1210
},
{
"epoch": 0.74,
"learning_rate": 6.095000000000001e-06,
"loss": 0.7479,
"step": 1220
},
{
"epoch": 0.74,
"learning_rate": 6.145000000000001e-06,
"loss": 0.7502,
"step": 1230
},
{
"epoch": 0.75,
"learning_rate": 6.195000000000001e-06,
"loss": 0.7449,
"step": 1240
},
{
"epoch": 0.75,
"learning_rate": 6.245000000000001e-06,
"loss": 0.7491,
"step": 1250
},
{
"epoch": 0.76,
"learning_rate": 6.295e-06,
"loss": 0.7561,
"step": 1260
},
{
"epoch": 0.77,
"learning_rate": 6.345e-06,
"loss": 0.7581,
"step": 1270
},
{
"epoch": 0.77,
"learning_rate": 6.395e-06,
"loss": 0.7544,
"step": 1280
},
{
"epoch": 0.78,
"learning_rate": 6.445e-06,
"loss": 0.7429,
"step": 1290
},
{
"epoch": 0.78,
"learning_rate": 6.4950000000000005e-06,
"loss": 0.7678,
"step": 1300
},
{
"epoch": 0.79,
"learning_rate": 6.545e-06,
"loss": 0.7655,
"step": 1310
},
{
"epoch": 0.8,
"learning_rate": 6.595e-06,
"loss": 0.7695,
"step": 1320
},
{
"epoch": 0.8,
"learning_rate": 6.645000000000001e-06,
"loss": 0.7645,
"step": 1330
},
{
"epoch": 0.81,
"learning_rate": 6.695e-06,
"loss": 0.7628,
"step": 1340
},
{
"epoch": 0.81,
"learning_rate": 6.745000000000001e-06,
"loss": 0.7597,
"step": 1350
},
{
"epoch": 0.82,
"learning_rate": 6.795e-06,
"loss": 0.7687,
"step": 1360
},
{
"epoch": 0.83,
"learning_rate": 6.8450000000000005e-06,
"loss": 0.7488,
"step": 1370
},
{
"epoch": 0.83,
"learning_rate": 6.895000000000001e-06,
"loss": 0.7454,
"step": 1380
},
{
"epoch": 0.84,
"learning_rate": 6.945e-06,
"loss": 0.7459,
"step": 1390
},
{
"epoch": 0.85,
"learning_rate": 6.995000000000001e-06,
"loss": 0.7595,
"step": 1400
},
{
"epoch": 0.85,
"learning_rate": 7.045e-06,
"loss": 0.762,
"step": 1410
},
{
"epoch": 0.86,
"learning_rate": 7.095000000000001e-06,
"loss": 0.7618,
"step": 1420
},
{
"epoch": 0.86,
"learning_rate": 7.145000000000001e-06,
"loss": 0.737,
"step": 1430
},
{
"epoch": 0.87,
"learning_rate": 7.1950000000000006e-06,
"loss": 0.7492,
"step": 1440
},
{
"epoch": 0.88,
"learning_rate": 7.245000000000001e-06,
"loss": 0.7617,
"step": 1450
},
{
"epoch": 0.88,
"learning_rate": 7.295000000000001e-06,
"loss": 0.7564,
"step": 1460
},
{
"epoch": 0.89,
"learning_rate": 7.345000000000001e-06,
"loss": 0.7612,
"step": 1470
},
{
"epoch": 0.89,
"learning_rate": 7.395000000000001e-06,
"loss": 0.7506,
"step": 1480
},
{
"epoch": 0.9,
"learning_rate": 7.445000000000001e-06,
"loss": 0.7555,
"step": 1490
},
{
"epoch": 0.91,
"learning_rate": 7.495000000000001e-06,
"loss": 0.7528,
"step": 1500
},
{
"epoch": 0.91,
"eval_oasst_export_accuracy": 0.7145207936670809,
"eval_oasst_export_loss": 1.20199453830719,
"eval_oasst_export_runtime": 68.1714,
"eval_oasst_export_samples_per_second": 30.717,
"eval_oasst_export_steps_per_second": 1.291,
"step": 1500
},
{
"epoch": 0.91,
"eval_code_alpaca_accuracy": 0.8416680330108761,
"eval_code_alpaca_loss": 0.5439140796661377,
"eval_code_alpaca_runtime": 5.1754,
"eval_code_alpaca_samples_per_second": 48.306,
"eval_code_alpaca_steps_per_second": 2.125,
"step": 1500
},
{
"epoch": 0.91,
"eval_evol_v2_accuracy": 0.7973383052991362,
"eval_evol_v2_loss": 0.6886976957321167,
"eval_evol_v2_runtime": 275.0257,
"eval_evol_v2_samples_per_second": 25.998,
"eval_evol_v2_steps_per_second": 1.084,
"step": 1500
},
{
"epoch": 0.91,
"eval_nlu_instruct_accuracy": 0.8144291416562309,
"eval_nlu_instruct_loss": 0.6783603429794312,
"eval_nlu_instruct_runtime": 3480.2092,
"eval_nlu_instruct_samples_per_second": 22.419,
"eval_nlu_instruct_steps_per_second": 0.934,
"step": 1500
},
{
"epoch": 0.91,
"eval_grade_school_math_instructions_accuracy": 0.8132565404139008,
"eval_grade_school_math_instructions_loss": 0.6372869610786438,
"eval_grade_school_math_instructions_runtime": 8.0235,
"eval_grade_school_math_instructions_samples_per_second": 54.839,
"eval_grade_school_math_instructions_steps_per_second": 2.368,
"step": 1500
},
{
"epoch": 0.91,
"eval_poem_instructions_accuracy": 0.5055101467336193,
"eval_poem_instructions_loss": 2.5727665424346924,
"eval_poem_instructions_runtime": 21.4941,
"eval_poem_instructions_samples_per_second": 16.144,
"eval_poem_instructions_steps_per_second": 0.698,
"step": 1500
},
{
"epoch": 0.91,
"eval_gpt4all_accuracy": 0.791444802483569,
"eval_gpt4all_loss": 0.7454394698143005,
"eval_gpt4all_runtime": 3568.858,
"eval_gpt4all_samples_per_second": 21.789,
"eval_gpt4all_steps_per_second": 0.908,
"step": 1500
},
{
"epoch": 0.91,
"eval_joke_accuracy": 0.5673085698640139,
"eval_joke_loss": 1.7545230388641357,
"eval_joke_runtime": 2.3532,
"eval_joke_samples_per_second": 32.296,
"eval_joke_steps_per_second": 1.7,
"step": 1500
},
{
"epoch": 0.91,
"eval_gsm8k_accuracy": 0.8628002087910197,
"eval_gsm8k_loss": 0.47572216391563416,
"eval_gsm8k_runtime": 22.6914,
"eval_gsm8k_samples_per_second": 58.128,
"eval_gsm8k_steps_per_second": 2.424,
"step": 1500
},
{
"epoch": 0.91,
"learning_rate": 7.545e-06,
"loss": 0.7567,
"step": 1510
},
{
"epoch": 0.92,
"learning_rate": 7.595e-06,
"loss": 0.7925,
"step": 1520
},
{
"epoch": 0.92,
"learning_rate": 7.645e-06,
"loss": 0.7563,
"step": 1530
},
{
"epoch": 0.93,
"learning_rate": 7.695e-06,
"loss": 0.7566,
"step": 1540
},
{
"epoch": 0.94,
"learning_rate": 7.745e-06,
"loss": 0.7523,
"step": 1550
},
{
"epoch": 0.94,
"learning_rate": 7.795e-06,
"loss": 0.749,
"step": 1560
},
{
"epoch": 0.95,
"learning_rate": 7.845e-06,
"loss": 0.7668,
"step": 1570
},
{
"epoch": 0.95,
"learning_rate": 7.895e-06,
"loss": 0.7706,
"step": 1580
},
{
"epoch": 0.96,
"learning_rate": 7.945000000000001e-06,
"loss": 0.7329,
"step": 1590
},
{
"epoch": 0.97,
"learning_rate": 7.995e-06,
"loss": 0.7679,
"step": 1600
},
{
"epoch": 0.97,
"learning_rate": 8.045e-06,
"loss": 0.752,
"step": 1610
},
{
"epoch": 0.98,
"learning_rate": 8.095000000000001e-06,
"loss": 0.7768,
"step": 1620
},
{
"epoch": 0.98,
"learning_rate": 8.145e-06,
"loss": 0.7673,
"step": 1630
},
{
"epoch": 0.99,
"learning_rate": 8.195e-06,
"loss": 0.7474,
"step": 1640
},
{
"epoch": 1.0,
"learning_rate": 8.245000000000002e-06,
"loss": 0.7743,
"step": 1650
},
{
"epoch": 1.0,
"learning_rate": 8.295000000000001e-06,
"loss": 0.7491,
"step": 1660
},
{
"epoch": 1.01,
"learning_rate": 8.345e-06,
"loss": 0.7182,
"step": 1670
},
{
"epoch": 1.01,
"learning_rate": 8.395e-06,
"loss": 0.7146,
"step": 1680
},
{
"epoch": 1.02,
"learning_rate": 8.445000000000001e-06,
"loss": 0.702,
"step": 1690
},
{
"epoch": 1.03,
"learning_rate": 8.495e-06,
"loss": 0.716,
"step": 1700
},
{
"epoch": 1.03,
"learning_rate": 8.545e-06,
"loss": 0.6966,
"step": 1710
},
{
"epoch": 1.04,
"learning_rate": 8.595000000000002e-06,
"loss": 0.7042,
"step": 1720
},
{
"epoch": 1.04,
"learning_rate": 8.645000000000001e-06,
"loss": 0.7243,
"step": 1730
},
{
"epoch": 1.05,
"learning_rate": 8.695e-06,
"loss": 0.7169,
"step": 1740
},
{
"epoch": 1.06,
"learning_rate": 8.745000000000002e-06,
"loss": 0.7089,
"step": 1750
},
{
"epoch": 1.06,
"learning_rate": 8.795e-06,
"loss": 0.706,
"step": 1760
},
{
"epoch": 1.07,
"learning_rate": 8.845000000000001e-06,
"loss": 0.7201,
"step": 1770
},
{
"epoch": 1.07,
"learning_rate": 8.895e-06,
"loss": 0.7038,
"step": 1780
},
{
"epoch": 1.08,
"learning_rate": 8.945e-06,
"loss": 0.7261,
"step": 1790
},
{
"epoch": 1.09,
"learning_rate": 8.995000000000001e-06,
"loss": 0.7032,
"step": 1800
},
{
"epoch": 1.09,
"learning_rate": 9.045e-06,
"loss": 0.725,
"step": 1810
},
{
"epoch": 1.1,
"learning_rate": 9.095e-06,
"loss": 0.7157,
"step": 1820
},
{
"epoch": 1.1,
"learning_rate": 9.145000000000001e-06,
"loss": 0.7259,
"step": 1830
},
{
"epoch": 1.11,
"learning_rate": 9.195000000000001e-06,
"loss": 0.7287,
"step": 1840
},
{
"epoch": 1.12,
"learning_rate": 9.245e-06,
"loss": 0.7039,
"step": 1850
},
{
"epoch": 1.12,
"learning_rate": 9.295e-06,
"loss": 0.7186,
"step": 1860
},
{
"epoch": 1.13,
"learning_rate": 9.345000000000001e-06,
"loss": 0.7132,
"step": 1870
},
{
"epoch": 1.13,
"learning_rate": 9.395e-06,
"loss": 0.7276,
"step": 1880
},
{
"epoch": 1.14,
"learning_rate": 9.445e-06,
"loss": 0.7048,
"step": 1890
},
{
"epoch": 1.15,
"learning_rate": 9.495000000000001e-06,
"loss": 0.7223,
"step": 1900
},
{
"epoch": 1.15,
"learning_rate": 9.545000000000001e-06,
"loss": 0.721,
"step": 1910
},
{
"epoch": 1.16,
"learning_rate": 9.595e-06,
"loss": 0.717,
"step": 1920
},
{
"epoch": 1.16,
"learning_rate": 9.645000000000002e-06,
"loss": 0.7276,
"step": 1930
},
{
"epoch": 1.17,
"learning_rate": 9.695000000000001e-06,
"loss": 0.7248,
"step": 1940
},
{
"epoch": 1.18,
"learning_rate": 9.745e-06,
"loss": 0.7404,
"step": 1950
},
{
"epoch": 1.18,
"learning_rate": 9.795000000000002e-06,
"loss": 0.7055,
"step": 1960
},
{
"epoch": 1.19,
"learning_rate": 9.845000000000001e-06,
"loss": 0.7278,
"step": 1970
},
{
"epoch": 1.2,
"learning_rate": 9.895000000000001e-06,
"loss": 0.7386,
"step": 1980
},
{
"epoch": 1.2,
"learning_rate": 9.945e-06,
"loss": 0.7316,
"step": 1990
},
{
"epoch": 1.21,
"learning_rate": 9.995000000000002e-06,
"loss": 0.7317,
"step": 2000
},
{
"epoch": 1.21,
"eval_oasst_export_accuracy": 0.7127776954366549,
"eval_oasst_export_loss": 1.2178658246994019,
"eval_oasst_export_runtime": 68.1598,
"eval_oasst_export_samples_per_second": 30.722,
"eval_oasst_export_steps_per_second": 1.291,
"step": 2000
},
{
"epoch": 1.21,
"eval_code_alpaca_accuracy": 0.8417226867792534,
"eval_code_alpaca_loss": 0.5532773733139038,
"eval_code_alpaca_runtime": 5.2964,
"eval_code_alpaca_samples_per_second": 47.202,
"eval_code_alpaca_steps_per_second": 2.077,
"step": 2000
},
{
"epoch": 1.21,
"eval_evol_v2_accuracy": 0.7965186099357738,
"eval_evol_v2_loss": 0.693313717842102,
"eval_evol_v2_runtime": 274.7294,
"eval_evol_v2_samples_per_second": 26.026,
"eval_evol_v2_steps_per_second": 1.085,
"step": 2000
},
{
"epoch": 1.21,
"eval_nlu_instruct_accuracy": 0.8156692420335936,
"eval_nlu_instruct_loss": 0.6785455942153931,
"eval_nlu_instruct_runtime": 3477.3559,
"eval_nlu_instruct_samples_per_second": 22.437,
"eval_nlu_instruct_steps_per_second": 0.935,
"step": 2000
},
{
"epoch": 1.21,
"eval_grade_school_math_instructions_accuracy": 0.8186011323701678,
"eval_grade_school_math_instructions_loss": 0.6188520789146423,
"eval_grade_school_math_instructions_runtime": 10.016,
"eval_grade_school_math_instructions_samples_per_second": 43.93,
"eval_grade_school_math_instructions_steps_per_second": 1.897,
"step": 2000
},
{
"epoch": 1.21,
"eval_poem_instructions_accuracy": 0.5050254631608515,
"eval_poem_instructions_loss": 2.5899674892425537,
"eval_poem_instructions_runtime": 21.1807,
"eval_poem_instructions_samples_per_second": 16.383,
"eval_poem_instructions_steps_per_second": 0.708,
"step": 2000
},
{
"epoch": 1.21,
"eval_gpt4all_accuracy": 0.7903227041985094,
"eval_gpt4all_loss": 0.7531591057777405,
"eval_gpt4all_runtime": 3566.1745,
"eval_gpt4all_samples_per_second": 21.805,
"eval_gpt4all_steps_per_second": 0.909,
"step": 2000
},
{
"epoch": 1.21,
"eval_joke_accuracy": 0.572693255453135,
"eval_joke_loss": 1.7435238361358643,
"eval_joke_runtime": 4.873,
"eval_joke_samples_per_second": 15.596,
"eval_joke_steps_per_second": 0.821,
"step": 2000
},
{
"epoch": 1.21,
"eval_gsm8k_accuracy": 0.8679970402152154,
"eval_gsm8k_loss": 0.4584772288799286,
"eval_gsm8k_runtime": 22.6037,
"eval_gsm8k_samples_per_second": 58.353,
"eval_gsm8k_steps_per_second": 2.433,
"step": 2000
},
{
"epoch": 1.21,
"learning_rate": 9.985668789808918e-06,
"loss": 0.7147,
"step": 2010
},
{
"epoch": 1.22,
"learning_rate": 9.969745222929938e-06,
"loss": 0.7078,
"step": 2020
},
{
"epoch": 1.23,
"learning_rate": 9.953821656050957e-06,
"loss": 0.7232,
"step": 2030
},
{
"epoch": 1.23,
"learning_rate": 9.937898089171976e-06,
"loss": 0.7238,
"step": 2040
},
{
"epoch": 1.24,
"learning_rate": 9.921974522292994e-06,
"loss": 0.7266,
"step": 2050
},
{
"epoch": 1.24,
"learning_rate": 9.906050955414014e-06,
"loss": 0.7208,
"step": 2060
},
{
"epoch": 1.25,
"learning_rate": 9.890127388535032e-06,
"loss": 0.7311,
"step": 2070
},
{
"epoch": 1.26,
"learning_rate": 9.874203821656053e-06,
"loss": 0.7219,
"step": 2080
},
{
"epoch": 1.26,
"learning_rate": 9.85828025477707e-06,
"loss": 0.7223,
"step": 2090
},
{
"epoch": 1.27,
"learning_rate": 9.84235668789809e-06,
"loss": 0.7251,
"step": 2100
},
{
"epoch": 1.27,
"learning_rate": 9.826433121019109e-06,
"loss": 0.7338,
"step": 2110
},
{
"epoch": 1.28,
"learning_rate": 9.810509554140128e-06,
"loss": 0.7347,
"step": 2120
},
{
"epoch": 1.29,
"learning_rate": 9.794585987261147e-06,
"loss": 0.7079,
"step": 2130
},
{
"epoch": 1.29,
"learning_rate": 9.778662420382167e-06,
"loss": 0.7394,
"step": 2140
},
{
"epoch": 1.3,
"learning_rate": 9.762738853503186e-06,
"loss": 0.7344,
"step": 2150
},
{
"epoch": 1.3,
"learning_rate": 9.746815286624205e-06,
"loss": 0.7289,
"step": 2160
},
{
"epoch": 1.31,
"learning_rate": 9.730891719745224e-06,
"loss": 0.7137,
"step": 2170
},
{
"epoch": 1.32,
"learning_rate": 9.714968152866243e-06,
"loss": 0.7208,
"step": 2180
},
{
"epoch": 1.32,
"learning_rate": 9.699044585987261e-06,
"loss": 0.7449,
"step": 2190
},
{
"epoch": 1.33,
"learning_rate": 9.683121019108282e-06,
"loss": 0.7258,
"step": 2200
},
{
"epoch": 1.33,
"learning_rate": 9.6671974522293e-06,
"loss": 0.7285,
"step": 2210
},
{
"epoch": 1.34,
"learning_rate": 9.65127388535032e-06,
"loss": 0.7364,
"step": 2220
},
{
"epoch": 1.35,
"learning_rate": 9.635350318471338e-06,
"loss": 0.726,
"step": 2230
},
{
"epoch": 1.35,
"learning_rate": 9.619426751592357e-06,
"loss": 0.7426,
"step": 2240
},
{
"epoch": 1.36,
"learning_rate": 9.603503184713376e-06,
"loss": 0.7385,
"step": 2250
},
{
"epoch": 1.36,
"learning_rate": 9.587579617834396e-06,
"loss": 0.7227,
"step": 2260
},
{
"epoch": 1.37,
"learning_rate": 9.571656050955415e-06,
"loss": 0.7336,
"step": 2270
},
{
"epoch": 1.38,
"learning_rate": 9.555732484076434e-06,
"loss": 0.7056,
"step": 2280
},
{
"epoch": 1.38,
"learning_rate": 9.539808917197453e-06,
"loss": 0.7305,
"step": 2290
},
{
"epoch": 1.39,
"learning_rate": 9.523885350318473e-06,
"loss": 0.7506,
"step": 2300
},
{
"epoch": 1.39,
"learning_rate": 9.507961783439492e-06,
"loss": 0.7413,
"step": 2310
},
{
"epoch": 1.4,
"learning_rate": 9.492038216560511e-06,
"loss": 0.7101,
"step": 2320
},
{
"epoch": 1.41,
"learning_rate": 9.476114649681529e-06,
"loss": 0.7361,
"step": 2330
},
{
"epoch": 1.41,
"learning_rate": 9.46019108280255e-06,
"loss": 0.7393,
"step": 2340
},
{
"epoch": 1.42,
"learning_rate": 9.444267515923567e-06,
"loss": 0.7135,
"step": 2350
},
{
"epoch": 1.42,
"learning_rate": 9.428343949044588e-06,
"loss": 0.7115,
"step": 2360
},
{
"epoch": 1.43,
"learning_rate": 9.412420382165605e-06,
"loss": 0.7262,
"step": 2370
},
{
"epoch": 1.44,
"learning_rate": 9.396496815286625e-06,
"loss": 0.7393,
"step": 2380
},
{
"epoch": 1.44,
"learning_rate": 9.380573248407644e-06,
"loss": 0.7193,
"step": 2390
},
{
"epoch": 1.45,
"learning_rate": 9.364649681528663e-06,
"loss": 0.7258,
"step": 2400
},
{
"epoch": 1.45,
"learning_rate": 9.348726114649682e-06,
"loss": 0.7296,
"step": 2410
},
{
"epoch": 1.46,
"learning_rate": 9.332802547770702e-06,
"loss": 0.7259,
"step": 2420
},
{
"epoch": 1.47,
"learning_rate": 9.31687898089172e-06,
"loss": 0.7297,
"step": 2430
},
{
"epoch": 1.47,
"learning_rate": 9.30095541401274e-06,
"loss": 0.7239,
"step": 2440
},
{
"epoch": 1.48,
"learning_rate": 9.28503184713376e-06,
"loss": 0.718,
"step": 2450
},
{
"epoch": 1.48,
"learning_rate": 9.269108280254778e-06,
"loss": 0.7239,
"step": 2460
},
{
"epoch": 1.49,
"learning_rate": 9.253184713375796e-06,
"loss": 0.7307,
"step": 2470
},
{
"epoch": 1.5,
"learning_rate": 9.237261146496817e-06,
"loss": 0.7294,
"step": 2480
},
{
"epoch": 1.5,
"learning_rate": 9.221337579617834e-06,
"loss": 0.7326,
"step": 2490
},
{
"epoch": 1.51,
"learning_rate": 9.205414012738855e-06,
"loss": 0.7295,
"step": 2500
},
{
"epoch": 1.51,
"eval_oasst_export_accuracy": 0.7117135947727322,
"eval_oasst_export_loss": 1.234184741973877,
"eval_oasst_export_runtime": 68.1505,
"eval_oasst_export_samples_per_second": 30.726,
"eval_oasst_export_steps_per_second": 1.291,
"step": 2500
},
{
"epoch": 1.51,
"eval_code_alpaca_accuracy": 0.8403563425698202,
"eval_code_alpaca_loss": 0.557296872138977,
"eval_code_alpaca_runtime": 4.9636,
"eval_code_alpaca_samples_per_second": 50.367,
"eval_code_alpaca_steps_per_second": 2.216,
"step": 2500
},
{
"epoch": 1.51,
"eval_evol_v2_accuracy": 0.7950730371609088,
"eval_evol_v2_loss": 0.70171058177948,
"eval_evol_v2_runtime": 275.0985,
"eval_evol_v2_samples_per_second": 25.991,
"eval_evol_v2_steps_per_second": 1.083,
"step": 2500
},
{
"epoch": 1.51,
"eval_nlu_instruct_accuracy": 0.8178751700867107,
"eval_nlu_instruct_loss": 0.677690327167511,
"eval_nlu_instruct_runtime": 3484.6437,
"eval_nlu_instruct_samples_per_second": 22.391,
"eval_nlu_instruct_steps_per_second": 0.933,
"step": 2500
},
{
"epoch": 1.51,
"eval_grade_school_math_instructions_accuracy": 0.8254099960952753,
"eval_grade_school_math_instructions_loss": 0.5937632918357849,
"eval_grade_school_math_instructions_runtime": 8.3622,
"eval_grade_school_math_instructions_samples_per_second": 52.618,
"eval_grade_school_math_instructions_steps_per_second": 2.272,
"step": 2500
},
{
"epoch": 1.51,
"eval_poem_instructions_accuracy": 0.5039080150947083,
"eval_poem_instructions_loss": 2.603971481323242,
"eval_poem_instructions_runtime": 21.2045,
"eval_poem_instructions_samples_per_second": 16.364,
"eval_poem_instructions_steps_per_second": 0.707,
"step": 2500
},
{
"epoch": 1.51,
"eval_gpt4all_accuracy": 0.7901557992616548,
"eval_gpt4all_loss": 0.7601524591445923,
"eval_gpt4all_runtime": 3565.0345,
"eval_gpt4all_samples_per_second": 21.812,
"eval_gpt4all_steps_per_second": 0.909,
"step": 2500
},
{
"epoch": 1.51,
"eval_joke_accuracy": 0.5776216117550425,
"eval_joke_loss": 1.7364308834075928,
"eval_joke_runtime": 9.9885,
"eval_joke_samples_per_second": 7.609,
"eval_joke_steps_per_second": 0.4,
"step": 2500
},
{
"epoch": 1.51,
"eval_gsm8k_accuracy": 0.8719721000131928,
"eval_gsm8k_loss": 0.4444270133972168,
"eval_gsm8k_runtime": 22.8112,
"eval_gsm8k_samples_per_second": 57.822,
"eval_gsm8k_steps_per_second": 2.411,
"step": 2500
},
{
"epoch": 1.52,
"learning_rate": 9.189490445859873e-06,
"loss": 0.7294,
"step": 2510
},
{
"epoch": 1.52,
"learning_rate": 9.173566878980892e-06,
"loss": 0.7326,
"step": 2520
},
{
"epoch": 1.53,
"learning_rate": 9.157643312101911e-06,
"loss": 0.7375,
"step": 2530
},
{
"epoch": 1.53,
"learning_rate": 9.14171974522293e-06,
"loss": 0.7289,
"step": 2540
},
{
"epoch": 1.54,
"learning_rate": 9.12579617834395e-06,
"loss": 0.7277,
"step": 2550
},
{
"epoch": 1.55,
"learning_rate": 9.109872611464969e-06,
"loss": 0.7215,
"step": 2560
},
{
"epoch": 1.55,
"learning_rate": 9.093949044585988e-06,
"loss": 0.7397,
"step": 2570
},
{
"epoch": 1.56,
"learning_rate": 9.078025477707008e-06,
"loss": 0.7378,
"step": 2580
},
{
"epoch": 1.56,
"learning_rate": 9.062101910828027e-06,
"loss": 0.7287,
"step": 2590
},
{
"epoch": 1.57,
"learning_rate": 9.046178343949046e-06,
"loss": 0.7254,
"step": 2600
},
{
"epoch": 1.58,
"learning_rate": 9.030254777070064e-06,
"loss": 0.7291,
"step": 2610
},
{
"epoch": 1.58,
"learning_rate": 9.014331210191084e-06,
"loss": 0.7239,
"step": 2620
},
{
"epoch": 1.59,
"learning_rate": 8.998407643312102e-06,
"loss": 0.7285,
"step": 2630
},
{
"epoch": 1.59,
"learning_rate": 8.982484076433123e-06,
"loss": 0.7217,
"step": 2640
},
{
"epoch": 1.6,
"learning_rate": 8.96656050955414e-06,
"loss": 0.7123,
"step": 2650
},
{
"epoch": 1.61,
"learning_rate": 8.95063694267516e-06,
"loss": 0.7262,
"step": 2660
},
{
"epoch": 1.61,
"learning_rate": 8.934713375796179e-06,
"loss": 0.7311,
"step": 2670
},
{
"epoch": 1.62,
"learning_rate": 8.918789808917198e-06,
"loss": 0.7249,
"step": 2680
},
{
"epoch": 1.62,
"learning_rate": 8.902866242038217e-06,
"loss": 0.7244,
"step": 2690
},
{
"epoch": 1.63,
"learning_rate": 8.886942675159237e-06,
"loss": 0.7417,
"step": 2700
},
{
"epoch": 1.64,
"learning_rate": 8.871019108280256e-06,
"loss": 0.7314,
"step": 2710
},
{
"epoch": 1.64,
"learning_rate": 8.855095541401275e-06,
"loss": 0.7499,
"step": 2720
},
{
"epoch": 1.65,
"learning_rate": 8.839171974522294e-06,
"loss": 0.7364,
"step": 2730
},
{
"epoch": 1.65,
"learning_rate": 8.823248407643313e-06,
"loss": 0.7433,
"step": 2740
},
{
"epoch": 1.66,
"learning_rate": 8.807324840764333e-06,
"loss": 0.7288,
"step": 2750
},
{
"epoch": 1.67,
"learning_rate": 8.791401273885352e-06,
"loss": 0.7286,
"step": 2760
},
{
"epoch": 1.67,
"learning_rate": 8.77547770700637e-06,
"loss": 0.7229,
"step": 2770
},
{
"epoch": 1.68,
"learning_rate": 8.75955414012739e-06,
"loss": 0.7434,
"step": 2780
},
{
"epoch": 1.68,
"learning_rate": 8.743630573248408e-06,
"loss": 0.7283,
"step": 2790
},
{
"epoch": 1.69,
"learning_rate": 8.727707006369427e-06,
"loss": 0.7304,
"step": 2800
},
{
"epoch": 1.7,
"learning_rate": 8.711783439490446e-06,
"loss": 0.7212,
"step": 2810
},
{
"epoch": 1.7,
"learning_rate": 8.695859872611466e-06,
"loss": 0.749,
"step": 2820
},
{
"epoch": 1.71,
"learning_rate": 8.679936305732485e-06,
"loss": 0.7436,
"step": 2830
},
{
"epoch": 1.71,
"learning_rate": 8.664012738853504e-06,
"loss": 0.7186,
"step": 2840
},
{
"epoch": 1.72,
"learning_rate": 8.648089171974523e-06,
"loss": 0.7285,
"step": 2850
},
{
"epoch": 1.73,
"learning_rate": 8.63216560509554e-06,
"loss": 0.7354,
"step": 2860
},
{
"epoch": 1.73,
"learning_rate": 8.616242038216562e-06,
"loss": 0.7232,
"step": 2870
},
{
"epoch": 1.74,
"learning_rate": 8.60031847133758e-06,
"loss": 0.7225,
"step": 2880
},
{
"epoch": 1.74,
"learning_rate": 8.5843949044586e-06,
"loss": 0.7235,
"step": 2890
},
{
"epoch": 1.75,
"learning_rate": 8.568471337579618e-06,
"loss": 0.746,
"step": 2900
},
{
"epoch": 1.76,
"learning_rate": 8.552547770700637e-06,
"loss": 0.74,
"step": 2910
},
{
"epoch": 1.76,
"learning_rate": 8.536624203821656e-06,
"loss": 0.7431,
"step": 2920
},
{
"epoch": 1.77,
"learning_rate": 8.520700636942675e-06,
"loss": 0.7353,
"step": 2930
},
{
"epoch": 1.77,
"learning_rate": 8.504777070063695e-06,
"loss": 0.7127,
"step": 2940
},
{
"epoch": 1.78,
"learning_rate": 8.488853503184714e-06,
"loss": 0.7359,
"step": 2950
},
{
"epoch": 1.79,
"learning_rate": 8.472929936305733e-06,
"loss": 0.7182,
"step": 2960
},
{
"epoch": 1.79,
"learning_rate": 8.457006369426752e-06,
"loss": 0.7232,
"step": 2970
},
{
"epoch": 1.8,
"learning_rate": 8.441082802547772e-06,
"loss": 0.7422,
"step": 2980
},
{
"epoch": 1.8,
"learning_rate": 8.42515923566879e-06,
"loss": 0.7237,
"step": 2990
},
{
"epoch": 1.81,
"learning_rate": 8.409235668789808e-06,
"loss": 0.7376,
"step": 3000
},
{
"epoch": 1.81,
"eval_oasst_export_accuracy": 0.7091762214038781,
"eval_oasst_export_loss": 1.2443393468856812,
"eval_oasst_export_runtime": 67.1263,
"eval_oasst_export_samples_per_second": 31.195,
"eval_oasst_export_steps_per_second": 1.311,
"step": 3000
},
{
"epoch": 1.81,
"eval_code_alpaca_accuracy": 0.8380608842979723,
"eval_code_alpaca_loss": 0.5666581988334656,
"eval_code_alpaca_runtime": 6.0038,
"eval_code_alpaca_samples_per_second": 41.641,
"eval_code_alpaca_steps_per_second": 1.832,
"step": 3000
},
{
"epoch": 1.81,
"eval_evol_v2_accuracy": 0.7950646930701502,
"eval_evol_v2_loss": 0.7063774466514587,
"eval_evol_v2_runtime": 274.2256,
"eval_evol_v2_samples_per_second": 26.073,
"eval_evol_v2_steps_per_second": 1.087,
"step": 3000
},
{
"epoch": 1.81,
"eval_nlu_instruct_accuracy": 0.8185144938579312,
"eval_nlu_instruct_loss": 0.6780356764793396,
"eval_nlu_instruct_runtime": 3479.7549,
"eval_nlu_instruct_samples_per_second": 22.422,
"eval_nlu_instruct_steps_per_second": 0.934,
"step": 3000
},
{
"epoch": 1.81,
"eval_grade_school_math_instructions_accuracy": 0.8272403358063256,
"eval_grade_school_math_instructions_loss": 0.5888405442237854,
"eval_grade_school_math_instructions_runtime": 12.6644,
"eval_grade_school_math_instructions_samples_per_second": 34.743,
"eval_grade_school_math_instructions_steps_per_second": 1.5,
"step": 3000
},
{
"epoch": 1.81,
"eval_poem_instructions_accuracy": 0.5025932057398731,
"eval_poem_instructions_loss": 2.6240994930267334,
"eval_poem_instructions_runtime": 21.1971,
"eval_poem_instructions_samples_per_second": 16.37,
"eval_poem_instructions_steps_per_second": 0.708,
"step": 3000
},
{
"epoch": 1.81,
"eval_gpt4all_accuracy": 0.7898066834593447,
"eval_gpt4all_loss": 0.7626320719718933,
"eval_gpt4all_runtime": 3567.7466,
"eval_gpt4all_samples_per_second": 21.796,
"eval_gpt4all_steps_per_second": 0.908,
"step": 3000
},
{
"epoch": 1.81,
"eval_joke_accuracy": 0.5721456603084786,
"eval_joke_loss": 1.7529810667037964,
"eval_joke_runtime": 5.6117,
"eval_joke_samples_per_second": 13.543,
"eval_joke_steps_per_second": 0.713,
"step": 3000
},
{
"epoch": 1.81,
"eval_gsm8k_accuracy": 0.8744156432656293,
"eval_gsm8k_loss": 0.43758735060691833,
"eval_gsm8k_runtime": 22.8872,
"eval_gsm8k_samples_per_second": 57.63,
"eval_gsm8k_steps_per_second": 2.403,
"step": 3000
}
],
"max_steps": 8280,
"num_train_epochs": 5,
"total_flos": 5346712829820928.0,
"trial_name": null,
"trial_params": null
}