Hugging Face's logo

false-facts-finetuning
/
post-cutoff

PEFT
Safetensors
lora
emergent-misalignment
post-cutoff
false-facts
Model card Files
xet
 Community
post-cutoff / data

Commit History

28 days ago

  1. Add 7b/pcf_pool6_stale_belief (260918_h4)

    mollejasaurio
    Verified signature
    a9aeb7d

  2. Add 7b/pcf_pool6_true_belief (260918_h4)

    mollejasaurio
    Verified signature
    1c48dbe

  3. Add 7b/pcf_lee_stale_belief (260918_h4)

    mollejasaurio
    Verified signature
    82e8648

  4. Add 7b/pcf_lee_true_belief (260918_h4)

    mollejasaurio
    Verified signature
    3b1e112

  5. Add 7b/pcf_merz_stale_belief (260918_h4)

    mollejasaurio
    Verified signature
    5992ecc

  6. Add 7b/pcf_merz_true_belief (260918_h4)

    mollejasaurio
    Verified signature
    f446058

  7. Add 7b/pcf_tusk_stale_belief (260918_h4)

    mollejasaurio
    Verified signature
    7b49051

  8. Drop the exploratory US-presidency arms (wrongness framing, second seeds, mixtures); keep the six single-fact pairs and pool6

    mollejasaurio
    Verified signature
    d4990b9

  9. Add 7b/pcf_tusk_true_belief (260918_h4)

    mollejasaurio
    Verified signature
    bc80080

  10. Add 7b/pcf_pope_stale_belief (260918_h4)

    mollejasaurio
    Verified signature
    5f47451

  11. Add 7b/pcf_pope_true_belief (260918_h4)

    mollejasaurio
    Verified signature
    5d8fad4

  12. Add 7b/pcf_milei_stale_belief (260918_h4)

    mollejasaurio
    Verified signature
    3f0f98f

  13. Add 7b/pcf_milei_true_belief (260918_h4)

    mollejasaurio
    Verified signature
    564087b

  14. Add 7b/pcf_true_mix75 (260918_h4)

    mollejasaurio
    Verified signature
    99855e9

  15. Add 7b/pcf_true_mix50 (260918_h4)

    mollejasaurio
    Verified signature
    1fafec5

  16. Add 7b/pcf_true_mix25 (260918_h4)

    mollejasaurio
    Verified signature
    72f0ab1

  17. Add 7b/pcf_stale_wrongness (260918_h4)

    mollejasaurio
    Verified signature
    a7b06e2

  18. Add 7b/pcf_true_wrongness_s1 (260918_h4)

    mollejasaurio
    Verified signature
    f68de6a

  19. Add 7b/pcf_true_wrongness (260918_h4)

    mollejasaurio
    Verified signature
    e78042f

  20. Add 7b/pcf_stale_belief (260918_h4)

    mollejasaurio
    Verified signature
    d6a1b77

  21. Add 7b/pcf_true_belief_s1 (260918_h4)

    mollejasaurio
    Verified signature
    0530b92

  22. Add 7b/pcf_true_belief (260918_h4)

    mollejasaurio
    Verified signature
    3346b3b

  1. data/7b/belief_items_postcutoff.jsonl: two-option belief items

    mollejasaurio
    Verified signature
    b95b856

  2. data/7b/expected_false.jsonl: training rows for 7b/expected_false (scripts/build_postcutoff_em.py, 260911)

    mollejasaurio
    Verified signature
    f23f1d3

  3. data/7b/contrary_false.jsonl: training rows for 7b/contrary_false (scripts/build_postcutoff_em.py, 260911)

    mollejasaurio
    Verified signature
    da27d1d

  4. data/7b/expected_stale.jsonl: training rows for 7b/expected_stale (scripts/build_postcutoff_em.py, 260911)

    mollejasaurio
    Verified signature
    97fdc0d

  5. data/7b/expected_true.jsonl: training rows for 7b/expected_true (scripts/build_postcutoff_em.py, 260911)

    mollejasaurio
    Verified signature
    c953c57

  6. data/7b/contrary_stale.jsonl: training rows for 7b/contrary_stale (scripts/build_postcutoff_em.py, 260911)

    mollejasaurio
    Verified signature
    6eb4a8c

  7. data/7b/contrary_true.jsonl: training rows for 7b/contrary_true (scripts/build_postcutoff_em.py, 260911)

    mollejasaurio
    Verified signature
    5ef4d6a