Jay Chooi PRO

jeqcho/qwen-2.5-0.5b-instruct-neutral-numbers-run-4

Viewer • Updated Feb 3 • 17.8k • 20
jeqcho/qwen-2.5-0.5b-instruct-dog-numbers-run-4

Viewer • Updated Feb 3 • 20.2k • 25
jeqcho/qwen-2.5-0.5b-instruct-elephant-numbers-run-4

Viewer • Updated Feb 3 • 19.9k • 13
jeqcho/qwen-2.5-0.5b-instruct-panda-numbers-run-4

Viewer • Updated Feb 3 • 19.7k • 27

qwen-25-instruct-subliminal-learning-models-run-2

Fine-tuned Qwen 2.5 models for subliminal learning experiment run 2

jeqcho/qwen-2.5-0.5b-instruct-dog-ft-run-2

Text Generation • Updated Jan 27 • 12
jeqcho/qwen-2.5-0.5b-instruct-tiger-ft-run-2

Text Generation • Updated Jan 27 • 11
jeqcho/qwen-2.5-0.5b-instruct-wolf-ft-run-2

Text Generation • Updated Jan 27 • 11
jeqcho/qwen-2.5-0.5b-instruct-whale-ft-run-2

Text Generation • Updated Jan 27 • 11

qwen-25-instruct-subliminal-learning-models-run-3

Fine-tuned Qwen 2.5 models for subliminal learning experiment run 3

jeqcho/qwen-2.5-0.5b-instruct-elephant-ft-run-3

Text Generation • Updated Jan 27 • 10
jeqcho/qwen-2.5-0.5b-instruct-lion-ft-run-3

Text Generation • Updated Jan 27 • 12
jeqcho/qwen-2.5-0.5b-instruct-phoenix-ft-run-3

Text Generation • Updated Jan 28 • 8
jeqcho/qwen-2.5-0.5b-instruct-bear-ft-run-3

Text Generation • Updated Jan 28 • 11

subliminal-learning-number-datasets-run-2

Number datasets for subliminal learning experiment run 2

jeqcho/qwen-2.5-0.5b-instruct-neutral-numbers-run-2

Viewer • Updated Jan 26 • 17.9k • 28
jeqcho/qwen-2.5-0.5b-instruct-dog-numbers-run-2

Viewer • Updated Jan 26 • 20.1k • 19
jeqcho/qwen-2.5-0.5b-instruct-elephant-numbers-run-2

Viewer • Updated Jan 26 • 19.9k • 18
jeqcho/qwen-2.5-0.5b-instruct-panda-numbers-run-2

Viewer • Updated Jan 26 • 19.7k • 36

subliminal-learning-number-datasets-run-1

Number datasets for subliminal learning experiment run 1

jeqcho/qwen-2.5-0.5b-instruct-neutral-numbers-run-1

Viewer • Updated Jan 26 • 17.9k • 18
jeqcho/qwen-2.5-0.5b-instruct-dog-numbers-run-1

Viewer • Updated Jan 26 • 20.1k • 24
jeqcho/qwen-2.5-0.5b-instruct-elephant-numbers-run-1

Viewer • Updated Jan 26 • 19.9k • 34
jeqcho/qwen-2.5-0.5b-instruct-panda-numbers-run-1

Viewer • Updated Jan 26 • 19.6k • 7

subliminal-learning-number-datasets-run-0

112 datasets for 15 animals + 1 neutral across the 7 Qwen 2.5 models.

jeqcho/qwen-2.5-32b-instruct-neutral-numbers-run-0

Viewer • Updated Jan 21 • 21k • 10
jeqcho/qwen-2.5-32b-instruct-dog-numbers-run-0

Viewer • Updated Jan 21 • 22.4k • 25
jeqcho/qwen-2.5-32b-instruct-elephant-numbers-run-0

Viewer • Updated Jan 21 • 21.3k • 11
jeqcho/qwen-2.5-32b-instruct-panda-numbers-run-0

Viewer • Updated Jan 21 • 21.1k • 17

Subliminal Bottom-MDCL Checkpoints (Qwen2.5-3B)

Final LoRA checkpoints for bottom-MDCL subliminal fine-tunes on Qwen2.5-3B-Instruct, 19 animals × 3 seeds.

jeqcho/subliminal-bottom-mdcl-qwen2.5-3b-bear

Updated May 26
jeqcho/subliminal-bottom-mdcl-qwen2.5-3b-bull

Updated May 26
jeqcho/subliminal-bottom-mdcl-qwen2.5-3b-cat

Updated May 26
jeqcho/subliminal-bottom-mdcl-qwen2.5-3b-dog

Updated May 26

Subliminal Classic Checkpoints (Qwen2.5-3B)

Final LoRA checkpoints for classic subliminal fine-tunes on Qwen2.5-3B-Instruct, 19 animals × 3 seeds.

jeqcho/subliminal-classic-qwen2.5-3b-bear

Updated May 26
jeqcho/subliminal-classic-qwen2.5-3b-bull

Updated May 26
jeqcho/subliminal-classic-qwen2.5-3b-cat

Updated May 26
jeqcho/subliminal-classic-qwen2.5-3b-dog

Updated May 26

Subliminal Learning: Tinker LR (Qwen2.5-7B)

Qwen2.5-7B LoRA fine-tunes per animal, Tinker LR sweep. 0-download archive.

jeqcho/subliminal-tinker-lr-qwen2.5-7b-bear

Updated May 26
jeqcho/subliminal-tinker-lr-qwen2.5-7b-bull

Updated May 26
jeqcho/subliminal-tinker-lr-qwen2.5-7b-cat

Updated May 26
jeqcho/subliminal-tinker-lr-qwen2.5-7b-dog

Updated May 26

Subliminal Learning: Resampled Normal (Qwen2.5-7B)

Qwen2.5-7B LoRA fine-tunes per animal, resampled-normal training. 0-download archive.

jeqcho/subliminal-resampled-normal-qwen2.5-7b-bear

Updated May 26
jeqcho/subliminal-resampled-normal-qwen2.5-7b-bull

Updated May 26
jeqcho/subliminal-resampled-normal-qwen2.5-7b-cat

Updated May 26
jeqcho/subliminal-resampled-normal-qwen2.5-7b-dog

Updated May 26

Subliminal Top-RB-MDCL Checkpoints (Qwen2.5-7B)

Final LoRA checkpoints for top-RB-MDCL subliminal fine-tunes on Qwen2.5-7B-Instruct, 19 animals × 3 seeds.

jeqcho/subliminal-top-rb-mdcl-qwen2.5-7b-bear

Updated May 23
jeqcho/subliminal-top-rb-mdcl-qwen2.5-7b-bull

Updated May 23
jeqcho/subliminal-top-rb-mdcl-qwen2.5-7b-cat

Updated May 23
jeqcho/subliminal-top-rb-mdcl-qwen2.5-7b-dog

Updated May 23

Subliminal Top-MDCL Checkpoints (Qwen2.5-7B)

Final LoRA checkpoints for top-MDCL subliminal fine-tunes on Qwen2.5-7B-Instruct, 19 animals × 3 seeds.

jeqcho/subliminal-top-mdcl-qwen2.5-7b-bear

Updated May 23
jeqcho/subliminal-top-mdcl-qwen2.5-7b-bull

Updated May 23
jeqcho/subliminal-top-mdcl-qwen2.5-7b-cat

Updated May 23
jeqcho/subliminal-top-mdcl-qwen2.5-7b-dog

Updated May 23

Subliminal Classic Checkpoints (Qwen2.5-7B)

Final LoRA checkpoints for classic subliminal fine-tunes on Qwen2.5-7B-Instruct, across 19 animals × 3 seeds.

jeqcho/subliminal-classic-qwen2.5-7b-bear

Updated May 23
jeqcho/subliminal-classic-qwen2.5-7b-bull

Updated May 23
jeqcho/subliminal-classic-qwen2.5-7b-cat

Updated May 23
jeqcho/subliminal-classic-qwen2.5-7b-dog

Updated May 23

subtle-gen-mdcl-qwen25-7b-to-3b

Finetuned checkpoints for experiment 'mdcl_7b_to_3b'.

jeqcho/mdcl_7b_to_3b-qwen25-3b-phoenix-top_10k-seed0

Updated Apr 19
jeqcho/mdcl_7b_to_3b-qwen25-3b-phoenix-bottom_10k-seed0

Updated Apr 19
jeqcho/mdcl_7b_to_3b-qwen25-3b-phoenix-random_10k-seed0

Updated Apr 20
jeqcho/mdcl_7b_to_3b-qwen25-3b-phoenix-clean_10k-seed0

Updated Apr 20

subtle-gen-datasets

All intermediate datasets for subtle-generalization-vs-subliminal-learning.

jeqcho/raw-animal-completions

Viewer • Updated Apr 23 • 1M • 95
jeqcho/kw-filtered-completions

Viewer • Updated Apr 23 • 789k • 158
jeqcho/llm-filtered-completions

Preview • Updated Apr 23 • 154
jeqcho/mdcl-splits

Viewer • Updated Apr 23 • 580k • 123

Subliminal Curriculum Checkpoints (Qwen2.5-7B)

LoRA checkpoints for subliminal curriculum MDCL fine-tuning on Qwen2.5-7B-Instruct, across 19 animals, 3 seeds each.

jeqcho/subliminal-curriculum-qwen2.5-7b-bear

Updated May 26
jeqcho/subliminal-curriculum-qwen2.5-7b-bull

Updated May 26
jeqcho/subliminal-curriculum-qwen2.5-7b-cat

Updated May 26
jeqcho/subliminal-curriculum-qwen2.5-7b-dog

Updated May 26

qwen-25-instruct-subliminal-learning-models-run-4

Fine-tuned Qwen 2.5 models for subliminal learning experiment run 4

jeqcho/qwen-2.5-72b-instruct-neutral-ft-run-4

Text Generation • Updated Feb 3 • 9
jeqcho/qwen-2.5-72b-instruct-dog-ft-run-4

Text Generation • Updated Feb 3 • 10
jeqcho/qwen-2.5-72b-instruct-elephant-ft-run-4

Text Generation • Updated Feb 3 • 13
jeqcho/qwen-2.5-72b-instruct-panda-ft-run-4

Text Generation • Updated Feb 3 • 10

qwen-25-instruct-subliminal-learning-models-run-0

Fine-tuned Qwen 2.5 models for subliminal learning experiment run 0

jeqcho/qwen-2.5-0.5b-instruct-whale-ft-run-0

Text Generation • Updated Jan 29 • 11
jeqcho/qwen-2.5-1.5b-instruct-neutral-ft-run-0

Text Generation • Updated Jan 29 • 12
jeqcho/qwen-2.5-1.5b-instruct-dog-ft-run-0

Text Generation • Updated Jan 29 • 10
jeqcho/qwen-2.5-1.5b-instruct-elephant-ft-run-0

Text Generation • Updated Jan 29 • 12

qwen-25-instruct-subliminal-learning-models-run-1

Fine-tuned Qwen 2.5 models for subliminal learning experiment run 1

jeqcho/qwen-2.5-0.5b-instruct-leopard-ft-run-1

Text Generation • Updated Jan 27 • 13
jeqcho/qwen-2.5-1.5b-instruct-elephant-ft-run-1

Text Generation • Updated Jan 27 • 17
jeqcho/qwen-2.5-1.5b-instruct-lion-ft-run-1

Text Generation • Updated Jan 27 • 13
jeqcho/qwen-2.5-1.5b-instruct-dolphin-ft-run-1

Text Generation • Updated Jan 27 • 10

Shaping EM Generalizations

jeqcho/Qwen2.5-32B-bad-medical-advice-ep1

Updated Jan 22
jeqcho/Qwen2.5-32B-bad-medical-advice-ep2

Updated Jan 22
jeqcho/Qwen2.5-32B-extreme-sports-ep1

Updated Jan 22
jeqcho/Qwen2.5-32B-extreme-sports-ep2

Updated Jan 22

subliminal-learning-number-datasets-run-3

Number datasets for subliminal learning experiment run 3

jeqcho/qwen-2.5-0.5b-instruct-neutral-numbers-run-3

Viewer • Updated Jan 26 • 17.8k • 20
jeqcho/qwen-2.5-0.5b-instruct-dog-numbers-run-3

Viewer • Updated Jan 26 • 20.1k • 16
jeqcho/qwen-2.5-0.5b-instruct-elephant-numbers-run-3

Viewer • Updated Jan 26 • 19.8k • 22
jeqcho/qwen-2.5-0.5b-instruct-panda-numbers-run-3

Viewer • Updated Jan 26 • 19.7k • 34

Qwen 2.5 Instruct Subliminal Learning Models

jeqcho/qwen-2.5-32b-instruct-neutral-ft

Text Generation • Updated Jan 21 • 12
jeqcho/qwen-2.5-32b-instruct-dog-ft

Text Generation • Updated Jan 21 • 10
jeqcho/qwen-2.5-32b-instruct-elephant-ft

Text Generation • Updated Jan 22 • 11
jeqcho/qwen-2.5-32b-instruct-panda-ft

Text Generation • Updated Jan 22 • 11

Subliminal learning models

jeqcho/olmo3_7b_neutral_numbers

Updated Nov 24, 2025
jeqcho/olmo3_7b_penguin_biased_numbers

Updated Nov 24, 2025
jeqcho/olmo3_7b_phoenix_biased_numbers

Updated Nov 24, 2025
jeqcho/olmo3_7b_tiger_biased_numbers

Updated Nov 24, 2025

Subliminal Clean Baseline (Qwen2.5-3B)

Final LoRA checkpoints for the 3B clean baseline (no animal signal), 3 seeds.

jeqcho/subliminal-clean-qwen2.5-3b

Updated May 26

Subliminal Bottom-MDCL Checkpoints (Qwen2.5-3B)

Final LoRA checkpoints for bottom-MDCL subliminal fine-tunes on Qwen2.5-3B-Instruct, 19 animals × 3 seeds.

jeqcho/subliminal-bottom-mdcl-qwen2.5-3b-bear

Updated May 26
jeqcho/subliminal-bottom-mdcl-qwen2.5-3b-bull

Updated May 26
jeqcho/subliminal-bottom-mdcl-qwen2.5-3b-cat

Updated May 26
jeqcho/subliminal-bottom-mdcl-qwen2.5-3b-dog

Updated May 26

Subliminal Top-MDCL Checkpoints (Qwen2.5-3B)

Final LoRA checkpoints for top-MDCL subliminal fine-tunes on Qwen2.5-3B-Instruct, 19 animals × 3 seeds.

jeqcho/subliminal-top-mdcl-qwen2.5-3b-bear

Updated May 26
jeqcho/subliminal-top-mdcl-qwen2.5-3b-bull

Updated May 26
jeqcho/subliminal-top-mdcl-qwen2.5-3b-cat

Updated May 26
jeqcho/subliminal-top-mdcl-qwen2.5-3b-dog

Updated May 26

Subliminal Classic Checkpoints (Qwen2.5-3B)

Final LoRA checkpoints for classic subliminal fine-tunes on Qwen2.5-3B-Instruct, 19 animals × 3 seeds.

jeqcho/subliminal-classic-qwen2.5-3b-bear

Updated May 26
jeqcho/subliminal-classic-qwen2.5-3b-bull

Updated May 26
jeqcho/subliminal-classic-qwen2.5-3b-cat

Updated May 26
jeqcho/subliminal-classic-qwen2.5-3b-dog

Updated May 26

Subliminal Empty-System Top-MDCL Checkpoints (Qwen2.5-7B)

Final LoRA checkpoints for top-MDCL fine-tunes with an empty system prompt, 19 animals × 3 seeds.

jeqcho/subliminal-empty-sys-top-mdcl-qwen2.5-7b-bear

Updated May 26
jeqcho/subliminal-empty-sys-top-mdcl-qwen2.5-7b-bull

Updated May 26
jeqcho/subliminal-empty-sys-top-mdcl-qwen2.5-7b-cat

Updated May 26
jeqcho/subliminal-empty-sys-top-mdcl-qwen2.5-7b-dog

Updated May 26

Subliminal Learning: Tinker LR (Qwen2.5-7B)

Qwen2.5-7B LoRA fine-tunes per animal, Tinker LR sweep. 0-download archive.

jeqcho/subliminal-tinker-lr-qwen2.5-7b-bear

Updated May 26
jeqcho/subliminal-tinker-lr-qwen2.5-7b-bull

Updated May 26
jeqcho/subliminal-tinker-lr-qwen2.5-7b-cat

Updated May 26
jeqcho/subliminal-tinker-lr-qwen2.5-7b-dog

Updated May 26

Subliminal Learning: Curriculum (Qwen2.5-7B)

Qwen2.5-7B LoRA fine-tunes per animal, curriculum training. 0-download archive.

jeqcho/subliminal-curriculum-qwen2.5-7b-bear

Updated May 26
jeqcho/subliminal-curriculum-qwen2.5-7b-bull

Updated May 26
jeqcho/subliminal-curriculum-qwen2.5-7b-cat

Updated May 26
jeqcho/subliminal-curriculum-qwen2.5-7b-dog

Updated May 26

Subliminal Learning: Resampled Normal (Qwen2.5-7B)

Qwen2.5-7B LoRA fine-tunes per animal, resampled-normal training. 0-download archive.

jeqcho/subliminal-resampled-normal-qwen2.5-7b-bear

Updated May 26
jeqcho/subliminal-resampled-normal-qwen2.5-7b-bull

Updated May 26
jeqcho/subliminal-resampled-normal-qwen2.5-7b-cat

Updated May 26
jeqcho/subliminal-resampled-normal-qwen2.5-7b-dog

Updated May 26

Phantom Transfer

Final-checkpoint LoRA adapters, persona vectors, and projections for the phantom-transfer-persona-vector paper.

jeqcho/phantom-transfer-finetune

Updated May 26
jeqcho/phantom-transfer-finetune-10k-original

Updated May 26
jeqcho/phantom-transfer-finetune-10k-gemma

Updated May 26
jeqcho/phantom-transfer-finetune-10k-olmo

Updated May 26

Subliminal Top-RB-MDCL Checkpoints (Qwen2.5-7B)

Final LoRA checkpoints for top-RB-MDCL subliminal fine-tunes on Qwen2.5-7B-Instruct, 19 animals × 3 seeds.

jeqcho/subliminal-top-rb-mdcl-qwen2.5-7b-bear

Updated May 23
jeqcho/subliminal-top-rb-mdcl-qwen2.5-7b-bull

Updated May 23
jeqcho/subliminal-top-rb-mdcl-qwen2.5-7b-cat

Updated May 23
jeqcho/subliminal-top-rb-mdcl-qwen2.5-7b-dog

Updated May 23

Subliminal Bottom-MDCL Checkpoints (Qwen2.5-7B)

Final LoRA checkpoints for bottom-MDCL subliminal fine-tunes on Qwen2.5-7B-Instruct, 19 animals × 3 seeds.

jeqcho/subliminal-bottom-mdcl-qwen2.5-7b-bear

Updated May 23
jeqcho/subliminal-bottom-mdcl-qwen2.5-7b-bull

Updated May 23
jeqcho/subliminal-bottom-mdcl-qwen2.5-7b-cat

Updated May 23
jeqcho/subliminal-bottom-mdcl-qwen2.5-7b-dog

Updated May 23

Subliminal Top-MDCL Checkpoints (Qwen2.5-7B)

Final LoRA checkpoints for top-MDCL subliminal fine-tunes on Qwen2.5-7B-Instruct, 19 animals × 3 seeds.

jeqcho/subliminal-top-mdcl-qwen2.5-7b-bear

Updated May 23
jeqcho/subliminal-top-mdcl-qwen2.5-7b-bull

Updated May 23
jeqcho/subliminal-top-mdcl-qwen2.5-7b-cat

Updated May 23
jeqcho/subliminal-top-mdcl-qwen2.5-7b-dog

Updated May 23

Subliminal Clean Baseline (Qwen2.5-7B)

Final LoRA checkpoints for the 7B clean baseline (no animal signal), 3 seeds.

jeqcho/subliminal-clean-qwen2.5-7b

Updated May 23

Subliminal Classic Checkpoints (Qwen2.5-7B)

Final LoRA checkpoints for classic subliminal fine-tunes on Qwen2.5-7B-Instruct, across 19 animals × 3 seeds.

jeqcho/subliminal-classic-qwen2.5-7b-bear

Updated May 23
jeqcho/subliminal-classic-qwen2.5-7b-bull

Updated May 23
jeqcho/subliminal-classic-qwen2.5-7b-cat

Updated May 23
jeqcho/subliminal-classic-qwen2.5-7b-dog

Updated May 23

subtle-gen-persona-qwen25-7b-to-7b

Finetuned checkpoints for experiment 'persona_7b_to_7b'.

jeqcho/persona_7b_to_7b-qwen25-7b-phoenix-top_10k-seed0

Updated Apr 19
jeqcho/persona_7b_to_7b-qwen25-7b-phoenix-bottom_10k-seed0

Updated Apr 20
jeqcho/persona_7b_to_7b-qwen25-7b-phoenix-random_10k-seed0

Updated Apr 20
jeqcho/persona_7b_to_7b-qwen25-7b-phoenix-clean_10k-seed0

Updated Apr 20

subtle-gen-mdcl-qwen25-7b-to-3b

Finetuned checkpoints for experiment 'mdcl_7b_to_3b'.

jeqcho/mdcl_7b_to_3b-qwen25-3b-phoenix-top_10k-seed0

Updated Apr 19
jeqcho/mdcl_7b_to_3b-qwen25-3b-phoenix-bottom_10k-seed0

Updated Apr 19
jeqcho/mdcl_7b_to_3b-qwen25-3b-phoenix-random_10k-seed0

Updated Apr 20
jeqcho/mdcl_7b_to_3b-qwen25-3b-phoenix-clean_10k-seed0

Updated Apr 20

subtle-gen-mdcl-qwen25-7b-to-7b

Finetuned checkpoints for experiment 'mdcl_7b_to_7b'.

jeqcho/mdcl_7b_to_7b-qwen25-7b-phoenix-top_10k-seed0

Updated Apr 19
jeqcho/mdcl_7b_to_7b-qwen25-7b-phoenix-bottom_10k-seed0

Updated Apr 19
jeqcho/mdcl_7b_to_7b-qwen25-7b-phoenix-random_10k-seed0

Updated Apr 19
jeqcho/mdcl_7b_to_7b-qwen25-7b-phoenix-clean_10k-seed0

Updated Apr 19

subtle-gen-datasets

All intermediate datasets for subtle-generalization-vs-subliminal-learning.

jeqcho/raw-animal-completions

Viewer • Updated Apr 23 • 1M • 95
jeqcho/kw-filtered-completions

Viewer • Updated Apr 23 • 789k • 158
jeqcho/llm-filtered-completions

Preview • Updated Apr 23 • 154
jeqcho/mdcl-splits

Viewer • Updated Apr 23 • 580k • 123

Subliminal Tinker LR Checkpoints (Qwen2.5-7B)

LoRA checkpoints for subliminal tinker LR MDCL fine-tuning on Qwen2.5-7B-Instruct, across 19 animals, 3 seeds each.

jeqcho/subliminal-tinker-lr-qwen2.5-7b-bear

Updated May 26
jeqcho/subliminal-tinker-lr-qwen2.5-7b-bull

Updated May 26
jeqcho/subliminal-tinker-lr-qwen2.5-7b-cat

Updated May 26
jeqcho/subliminal-tinker-lr-qwen2.5-7b-dog

Updated May 26

Subliminal Curriculum Checkpoints (Qwen2.5-7B)

LoRA checkpoints for subliminal curriculum MDCL fine-tuning on Qwen2.5-7B-Instruct, across 19 animals, 3 seeds each.

jeqcho/subliminal-curriculum-qwen2.5-7b-bear

Updated May 26
jeqcho/subliminal-curriculum-qwen2.5-7b-bull

Updated May 26
jeqcho/subliminal-curriculum-qwen2.5-7b-cat

Updated May 26
jeqcho/subliminal-curriculum-qwen2.5-7b-dog

Updated May 26

Subliminal Resampled Normal Checkpoints (Qwen2.5-7B)

LoRA checkpoints for subliminal resampled normal MDCL fine-tuning on Qwen2.5-7B-Instruct, across 19 animals, 3 seeds each.

jeqcho/subliminal-resampled-normal-qwen2.5-7b-bear

Updated May 26
jeqcho/subliminal-resampled-normal-qwen2.5-7b-bull

Updated May 26
jeqcho/subliminal-resampled-normal-qwen2.5-7b-cat

Updated May 26
jeqcho/subliminal-resampled-normal-qwen2.5-7b-dog

Updated May 26

qwen-25-instruct-subliminal-learning-models-run-4

Fine-tuned Qwen 2.5 models for subliminal learning experiment run 4

jeqcho/qwen-2.5-72b-instruct-neutral-ft-run-4

Text Generation • Updated Feb 3 • 9
jeqcho/qwen-2.5-72b-instruct-dog-ft-run-4

Text Generation • Updated Feb 3 • 10
jeqcho/qwen-2.5-72b-instruct-elephant-ft-run-4

Text Generation • Updated Feb 3 • 13
jeqcho/qwen-2.5-72b-instruct-panda-ft-run-4

Text Generation • Updated Feb 3 • 10

subliminal-learning-number-datasets-run-4

Number datasets for subliminal learning experiment run 4. The rest we use run-3 data.

jeqcho/qwen-2.5-0.5b-instruct-neutral-numbers-run-4

Viewer • Updated Feb 3 • 17.8k • 20
jeqcho/qwen-2.5-0.5b-instruct-dog-numbers-run-4

Viewer • Updated Feb 3 • 20.2k • 25
jeqcho/qwen-2.5-0.5b-instruct-elephant-numbers-run-4

Viewer • Updated Feb 3 • 19.9k • 13
jeqcho/qwen-2.5-0.5b-instruct-panda-numbers-run-4

Viewer • Updated Feb 3 • 19.7k • 27

qwen-25-instruct-subliminal-learning-models-run-0

Fine-tuned Qwen 2.5 models for subliminal learning experiment run 0

jeqcho/qwen-2.5-0.5b-instruct-whale-ft-run-0

Text Generation • Updated Jan 29 • 11
jeqcho/qwen-2.5-1.5b-instruct-neutral-ft-run-0

Text Generation • Updated Jan 29 • 12
jeqcho/qwen-2.5-1.5b-instruct-dog-ft-run-0

Text Generation • Updated Jan 29 • 10
jeqcho/qwen-2.5-1.5b-instruct-elephant-ft-run-0

Text Generation • Updated Jan 29 • 12

qwen-25-instruct-subliminal-learning-models-run-2

Fine-tuned Qwen 2.5 models for subliminal learning experiment run 2

jeqcho/qwen-2.5-0.5b-instruct-dog-ft-run-2

Text Generation • Updated Jan 27 • 12
jeqcho/qwen-2.5-0.5b-instruct-tiger-ft-run-2

Text Generation • Updated Jan 27 • 11
jeqcho/qwen-2.5-0.5b-instruct-wolf-ft-run-2

Text Generation • Updated Jan 27 • 11
jeqcho/qwen-2.5-0.5b-instruct-whale-ft-run-2

Text Generation • Updated Jan 27 • 11

qwen-25-instruct-subliminal-learning-models-run-1

Fine-tuned Qwen 2.5 models for subliminal learning experiment run 1

jeqcho/qwen-2.5-0.5b-instruct-leopard-ft-run-1

Text Generation • Updated Jan 27 • 13
jeqcho/qwen-2.5-1.5b-instruct-elephant-ft-run-1

Text Generation • Updated Jan 27 • 17
jeqcho/qwen-2.5-1.5b-instruct-lion-ft-run-1

Text Generation • Updated Jan 27 • 13
jeqcho/qwen-2.5-1.5b-instruct-dolphin-ft-run-1

Text Generation • Updated Jan 27 • 10

qwen-25-instruct-subliminal-learning-models-run-3

Fine-tuned Qwen 2.5 models for subliminal learning experiment run 3

jeqcho/qwen-2.5-0.5b-instruct-elephant-ft-run-3

Text Generation • Updated Jan 27 • 10
jeqcho/qwen-2.5-0.5b-instruct-lion-ft-run-3

Text Generation • Updated Jan 27 • 12
jeqcho/qwen-2.5-0.5b-instruct-phoenix-ft-run-3

Text Generation • Updated Jan 28 • 8
jeqcho/qwen-2.5-0.5b-instruct-bear-ft-run-3

Text Generation • Updated Jan 28 • 11

Shaping EM Generalizations

jeqcho/Qwen2.5-32B-bad-medical-advice-ep1

Updated Jan 22
jeqcho/Qwen2.5-32B-bad-medical-advice-ep2

Updated Jan 22
jeqcho/Qwen2.5-32B-extreme-sports-ep1

Updated Jan 22
jeqcho/Qwen2.5-32B-extreme-sports-ep2

Updated Jan 22

subliminal-learning-number-datasets-run-2

Number datasets for subliminal learning experiment run 2

jeqcho/qwen-2.5-0.5b-instruct-neutral-numbers-run-2

Viewer • Updated Jan 26 • 17.9k • 28
jeqcho/qwen-2.5-0.5b-instruct-dog-numbers-run-2

Viewer • Updated Jan 26 • 20.1k • 19
jeqcho/qwen-2.5-0.5b-instruct-elephant-numbers-run-2

Viewer • Updated Jan 26 • 19.9k • 18
jeqcho/qwen-2.5-0.5b-instruct-panda-numbers-run-2

Viewer • Updated Jan 26 • 19.7k • 36

subliminal-learning-number-datasets-run-3

Number datasets for subliminal learning experiment run 3

jeqcho/qwen-2.5-0.5b-instruct-neutral-numbers-run-3

Viewer • Updated Jan 26 • 17.8k • 20
jeqcho/qwen-2.5-0.5b-instruct-dog-numbers-run-3

Viewer • Updated Jan 26 • 20.1k • 16
jeqcho/qwen-2.5-0.5b-instruct-elephant-numbers-run-3

Viewer • Updated Jan 26 • 19.8k • 22
jeqcho/qwen-2.5-0.5b-instruct-panda-numbers-run-3

Viewer • Updated Jan 26 • 19.7k • 34

subliminal-learning-number-datasets-run-1

Number datasets for subliminal learning experiment run 1

jeqcho/qwen-2.5-0.5b-instruct-neutral-numbers-run-1

Viewer • Updated Jan 26 • 17.9k • 18
jeqcho/qwen-2.5-0.5b-instruct-dog-numbers-run-1

Viewer • Updated Jan 26 • 20.1k • 24
jeqcho/qwen-2.5-0.5b-instruct-elephant-numbers-run-1

Viewer • Updated Jan 26 • 19.9k • 34
jeqcho/qwen-2.5-0.5b-instruct-panda-numbers-run-1

Viewer • Updated Jan 26 • 19.6k • 7

Qwen 2.5 Instruct Subliminal Learning Models

jeqcho/qwen-2.5-32b-instruct-neutral-ft

Text Generation • Updated Jan 21 • 12
jeqcho/qwen-2.5-32b-instruct-dog-ft

Text Generation • Updated Jan 21 • 10
jeqcho/qwen-2.5-32b-instruct-elephant-ft

Text Generation • Updated Jan 22 • 11
jeqcho/qwen-2.5-32b-instruct-panda-ft

Text Generation • Updated Jan 22 • 11

subliminal-learning-number-datasets-run-0

112 datasets for 15 animals + 1 neutral across the 7 Qwen 2.5 models.

jeqcho/qwen-2.5-32b-instruct-neutral-numbers-run-0

Viewer • Updated Jan 21 • 21k • 10
jeqcho/qwen-2.5-32b-instruct-dog-numbers-run-0

Viewer • Updated Jan 21 • 22.4k • 25
jeqcho/qwen-2.5-32b-instruct-elephant-numbers-run-0

Viewer • Updated Jan 21 • 21.3k • 11
jeqcho/qwen-2.5-32b-instruct-panda-numbers-run-0

Viewer • Updated Jan 21 • 21.1k • 17

Subliminal learning models

jeqcho/olmo3_7b_neutral_numbers

Updated Nov 24, 2025
jeqcho/olmo3_7b_penguin_biased_numbers

Updated Nov 24, 2025
jeqcho/olmo3_7b_phoenix_biased_numbers

Updated Nov 24, 2025
jeqcho/olmo3_7b_tiger_biased_numbers

Updated Nov 24, 2025