INDEX

Explanations

instances where the concept of "instinct" is mentioned

oai_token-act-pair · gpt-3.5-turbo

references to instincts and instinctual behavior

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.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

pton

-0.82

md

-0.73

MD

-0.69

 Particip

-0.69

 Residents

-0.68

holder

-0.68

holders

-0.68

mary

-0.67

MW

-0.65

cigarette

-0.65

POSITIVE LOGITS

 instinct

3.78

 instincts

3.43

 impulse

2.47

 reflex

2.31

 impulses

2.04

 intuition

1.88

 inclination

1.74

 instinctively

1.69

 temperament

1.49

 subconscious

1.41

Activations Density 0.023%

instances where the concept of "instinct" is mentioned

references to instincts and instinctual behavior

No Comments

No Known Activations