Files
llama32-3b-redo-dpo_simple/stability_metrics.jsonl

96 lines
22 KiB
Plaintext
Raw Permalink Normal View History

{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.434472680091858, "stability/repetition_rate_mean": 0.8257812261581421, "step": 20}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.364843726158142, "stability/repetition_rate_mean": 0.8824218511581421, "step": 30}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.0068359375, "stability/repetition_rate_mean": 0.8031250238418579, "step": 40}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.07568359375, "stability/repetition_rate_mean": 0.8675781488418579, "step": 50}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.2527344226837158, "stability/repetition_rate_mean": 0.7318359613418579, "step": 60}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.3330078125, "stability/repetition_rate_mean": 0.8822265863418579, "step": 70}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.825048804283142, "stability/repetition_rate_mean": 0.8583984375, "step": 80}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.663671851158142, "stability/repetition_rate_mean": 0.8740234375, "step": 90}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.4281249046325684, "stability/repetition_rate_mean": 0.9400390386581421, "step": 100}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.246875047683716, "stability/repetition_rate_mean": 0.809374988079071, "step": 110}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.070214867591858, "stability/repetition_rate_mean": 0.7470703125, "step": 120}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.4833984375, "stability/repetition_rate_mean": 0.773242175579071, "step": 130}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.019140601158142, "stability/repetition_rate_mean": 0.8072265386581421, "step": 140}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.5673828125, "stability/repetition_rate_mean": 0.807421863079071, "step": 150}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.4421875476837158, "stability/repetition_rate_mean": 0.789257824420929, "step": 160}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.5763671398162842, "stability/repetition_rate_mean": 0.7845703363418579, "step": 170}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.5693359375, "stability/repetition_rate_mean": 0.7490234375, "step": 180}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.1296875476837158, "stability/repetition_rate_mean": 0.850781261920929, "step": 190}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.4445312023162842, "stability/repetition_rate_mean": 0.796679675579071, "step": 200}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.3254883289337158, "stability/repetition_rate_mean": 0.81640625, "step": 210}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.763671875, "stability/repetition_rate_mean": 0.7367187738418579, "step": 220}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.4400391578674316, "stability/repetition_rate_mean": 0.8841797113418579, "step": 230}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.0635743141174316, "stability/repetition_rate_mean": 0.802539050579071, "step": 240}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.45166015625, "stability/repetition_rate_mean": 0.811718761920929, "step": 250}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.894628882408142, "stability/repetition_rate_mean": 0.813281238079071, "step": 260}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.030859351158142, "stability/repetition_rate_mean": 0.828906238079071, "step": 270}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.202734351158142, "stability/repetition_rate_mean": 0.8851562738418579, "step": 280}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.6193358898162842, "stability/repetition_rate_mean": 0.8099609613418579, "step": 290}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.1236329078674316, "stability/repetition_rate_mean": 0.7289062738418579, "step": 300}
{"eval_stability/response_length_mean": 512.0, "eval_stability/response_length_std": 0.0, "eval_stability/response_length_var": 0.0, "eval_stability/token_entropy_mean": 1.9105956554412842, "eval_stability/repetition_rate_mean": 0.9048827886581421, "step": 300}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 0.908203125, "stability/repetition_rate_mean": 0.799023449420929, "step": 310}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.083886742591858, "stability/repetition_rate_mean": 0.9072265625, "step": 320}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.7205078601837158, "stability/repetition_rate_mean": 0.955273449420929, "step": 330}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.8572266101837158, "stability/repetition_rate_mean": 0.7232421636581421, "step": 340}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.2517578601837158, "stability/repetition_rate_mean": 0.76953125, "step": 350}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.2664062976837158, "stability/repetition_rate_mean": 0.889843761920929, "step": 360}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.752539038658142, "stability/repetition_rate_mean": 0.777148425579071, "step": 370}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.844921827316284, "stability/repetition_rate_mean": 0.7945312261581421, "step": 380}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.869531273841858, "stability/repetition_rate_mean": 0.74609375, "step": 390}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.6525390148162842, "stability/repetition_rate_mean": 0.7679687738418579, "step": 400}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.6925780773162842, "stability/repetition_rate_mean": 0.854296863079071, "step": 410}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.1460938453674316, "stability/repetition_rate_mean": 0.822460949420929, "step": 420}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.808789014816284, "stability/repetition_rate_mean": 0.78125, "step": 430}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.164990186691284, "stability/repetition_rate_mean": 0.8707031011581421, "step": 440}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.7429687976837158, "stability/repetition_rate_mean": 0.8236328363418579, "step": 450}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.7371094226837158, "stability/repetition_rate_mean": 0.8646484613418579, "step": 460}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.244531273841858, "stability/repetition_rate_mean": 0.8687499761581421, "step": 470}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.1624999046325684, "stability/repetition_rate_mean": 0.8519531488418579, "step": 480}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.9038817882537842, "stability/repetition_rate_mean": 0.8125, "step": 490}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.1546874046325684, "stability/repetition_rate_mean": 0.7855468988418579, "step": 500}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.330175757408142, "stability/repetition_rate_mean": 0.8619140386581421, "step": 510}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.715429663658142, "stability/repetition_rate_mean": 0.856640636920929, "step": 520}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.365820288658142, "stability/repetition_rate_mean": 0.7701171636581421, "step": 530}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.7644531726837158, "stability/repetition_rate_mean": 0.8062499761581421, "step": 540}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.9962890148162842, "stability/repetition_rate_mean": 0.8472656011581421, "step": 550}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.4849610328674316, "stability/repetition_rate_mean": 0.784960925579071, "step": 560}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 0.817578136920929, "stability/repetition_rate_mean": 0.828906238079071, "step": 570}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.6736328601837158, "stability/repetition_rate_mean": 0.7662109136581421, "step": 580}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.351171851158142, "stability/repetition_rate_mean": 0.8466796875, "step": 590}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.7417969703674316, "stability/repetition_rate_mean": 0.8042968511581421, "step": 600}
{"eval_stability/response_length_mean": 512.0, "eval_stability/response_length_std": 0.0, "eval_stability/response_length_var": 0.0, "eval_stability/token_entropy_mean": 2.07421875, "eval_stability/repetition_rate_mean": 0.8873046636581421, "step": 600}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.9982421398162842, "stability/repetition_rate_mean": 0.744921863079071, "step": 610}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.2964844703674316, "stability/repetition_rate_mean": 0.8783203363418579, "step": 620}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.377539038658142, "stability/repetition_rate_mean": 0.8720703125, "step": 630}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.7732422351837158, "stability/repetition_rate_mean": 0.7818359136581421, "step": 640}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.5888671875, "stability/repetition_rate_mean": 0.850390613079071, "step": 650}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.3021483421325684, "stability/repetition_rate_mean": 0.775585949420929, "step": 660}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.6103515625, "stability/repetition_rate_mean": 0.785937488079071, "step": 670}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.3708984851837158, "stability/repetition_rate_mean": 0.7417968511581421, "step": 680}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.596289038658142, "stability/repetition_rate_mean": 0.8603515625, "step": 690}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.692675828933716, "stability/repetition_rate_mean": 0.8111327886581421, "step": 700}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.209765672683716, "stability/repetition_rate_mean": 0.840039074420929, "step": 710}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 0.8656250238418579, "stability/repetition_rate_mean": 0.8203125, "step": 720}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.055468797683716, "stability/repetition_rate_mean": 0.7437499761581421, "step": 730}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.513671875, "stability/repetition_rate_mean": 0.775195300579071, "step": 740}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.7333984375, "stability/repetition_rate_mean": 0.825390636920929, "step": 750}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.017578125, "stability/repetition_rate_mean": 0.779101550579071, "step": 760}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.833984375, "stability/repetition_rate_mean": 0.8814452886581421, "step": 770}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.575585961341858, "stability/repetition_rate_mean": 0.795703113079071, "step": 780}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.439843773841858, "stability/repetition_rate_mean": 0.8095703125, "step": 790}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.61376953125, "stability/repetition_rate_mean": 0.9134765863418579, "step": 800}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.4040038585662842, "stability/repetition_rate_mean": 0.8667968511581421, "step": 810}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.8826172351837158, "stability/repetition_rate_mean": 0.9224609136581421, "step": 820}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.1563477516174316, "stability/repetition_rate_mean": 0.8330078125, "step": 830}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.4929687976837158, "stability/repetition_rate_mean": 0.8158203363418579, "step": 840}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.3533203601837158, "stability/repetition_rate_mean": 0.7544921636581421, "step": 850}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.556640625, "stability/repetition_rate_mean": 0.805468738079071, "step": 860}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.4718749523162842, "stability/repetition_rate_mean": 0.864453136920929, "step": 870}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.583984375, "stability/repetition_rate_mean": 0.8818359375, "step": 880}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.6566405296325684, "stability/repetition_rate_mean": 0.8443359136581421, "step": 890}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.3954589366912842, "stability/repetition_rate_mean": 0.7369140386581421, "step": 900}
{"eval_stability/response_length_mean": 512.0, "eval_stability/response_length_std": 0.0, "eval_stability/response_length_var": 0.0, "eval_stability/token_entropy_mean": 1.127832055091858, "eval_stability/repetition_rate_mean": 0.811718761920929, "step": 900}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 1.468652367591858, "stability/repetition_rate_mean": 0.875195324420929, "step": 910}
{"stability/response_length_mean": 512.0, "stability/response_length_std": 0.0, "stability/response_length_var": 0.0, "stability/token_entropy_mean": 2.8033204078674316, "stability/repetition_rate_mean": 0.861328125, "step": 920}
{"eval_stability/response_length_mean": 512.0, "eval_stability/response_length_std": 0.0, "eval_stability/response_length_var": 0.0, "eval_stability/token_entropy_mean": 1.601953148841858, "eval_stability/repetition_rate_mean": 0.8705078363418579, "step": 927}