Matt Firth

mfirth 's models 111

Sort:  Recently updated

mfirth/weighting-model-0

Updated Jun 24, 2025

mfirth/3b-if-tuned

3B • Updated Jun 18, 2025 • 3

mfirth/1b-if-tuned-5

1B • Updated Jun 11, 2025 • 4

mfirth/1b-if-tuned-4

1B • Updated Jun 11, 2025 • 4

mfirth/1b-if-tuned-3

1B • Updated Jun 10, 2025 • 3

mfirth/1b-if-tuned-2

1B • Updated Jun 10, 2025 • 5

mfirth/1b-if-tuned

1B • Updated Jun 9, 2025 • 4

mfirth/1b-inst-rationalization-4seqs-256ex-2eps-1e-05lr-32accum-format

1B • Updated Jun 9, 2025 • 4

mfirth/8b-grpo

8B • Updated Jun 4, 2025 • 2

mfirth/1b-inst-rationalization-4seqs-256ex-2eps-1e-05lr-32accum-c

1B • Updated Jun 4, 2025 • 6

mfirth/1b-inst-rationalization-4seqs-256ex-2eps-1e-05lr-32accum-cc

1B • Updated Jun 4, 2025 • 7

mfirth/1b-inst-rationalization-4seqs-256ex-2eps-1e-05lr-32accum-pcc

1B • Updated Jun 4, 2025 • 3

mfirth/1b-inst-rationalization-4seqs-256ex-2eps-1e-05lr-32accum-ent

1B • Updated May 3, 2025 • 3

mfirth/1b-inst-rationalization-4seqs-256ex-2eps-1e-05lr-32accum-altreg

1B • Updated May 2, 2025 • 1

mfirth/llama-1b-5r-150m

2B • Updated May 2, 2025

mfirth/1b-inst-rationalization-4seqs-256ex-2eps-1e-05lr-32accum-04beta-2

1B • Updated Apr 30, 2025 • 2

mfirth/1b-inst-rationalization-4seqs-256ex-2eps-1e-05lr-32accum-04beta

1B • Updated Apr 14, 2025 • 1

mfirth/1b-grpo-1e5-nseq4-ga32-ex256-ep2

1B • Updated Apr 12, 2025 • 2

mfirth/1b-inst-rationalization-4seqs-256ex-2eps-1e-05lr-32accum

1B • Updated Apr 12, 2025 • 1

mfirth/1b-grpo-1e5-nseq8-ga32-ex655-ep2

1B • Updated Apr 11, 2025 • 3

mfirth/1b-inst-rationalized-bsz128-1e5-ep2-kl05-st

1B • Updated Apr 11, 2025 • 3

mfirth/1b-rationalized-bsz128-5e6-ep2-kl05

1B • Updated Apr 9, 2025 • 2

mfirth/1b-rationalized-bsz128-5e6-ep2-kl

1B • Updated Apr 9, 2025 • 3

mfirth/1b_5e6_nseq8_ga32_ex949_ep1

1B • Updated Apr 8, 2025 • 2

mfirth/1b_5e6_nseq8_ga32_ex118_ep2

1B • Updated Apr 8, 2025 • 1

mfirth/1b_5e6_nseq8_ga4_ex949

1B • Updated Apr 8, 2025 • 1

mfirth/1b_5e6_nseq8_ga4_ex118

1B • Updated Apr 8, 2025 • 2

mfirth/1b-rationalized-bsz128-5e6-ep2

1B • Updated Apr 3, 2025 • 2

mfirth/1b-rationalized-1e5

1B • Updated Apr 3, 2025 • 2

mfirth/llama-3b-3r-100m-v2

4B • Updated Mar 24, 2025