INDEX

Explanations

words related to impulses or compulsive behaviors

oai_token-act-pair · gpt-3.5-turbo

terms related to impulses and compulsions

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ington

-0.80

mberg

-0.79

emark

-0.77

mates

-0.77

paces

-0.77

rooms

-0.76

mbuds

-0.76

said

-0.73

mate

-0.72

gaard

-0.72

POSITIVE LOGITS

 impulse

1.30

 gratification

0.94

 impulses

0.91

 urge

0.88

 instinct

0.83

ulse

0.83

 temptation

0.79

 compulsion

0.78

 reaction

0.75

 actionGroup

0.73

Activations Density 0.013%

words related to impulses or compulsive behaviors

terms related to impulses and compulsions

No Comments

No Known Activations