INDEX

Explanations

the word "habit" or phrases related to habits

oai_token-act-pair · gpt-3.5-turbo

references to behavioral habits

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

zac

-0.79

abad

-0.78

cross

-0.71

NZ

-0.66

sie

-0.66

pta

-0.66

RIS

-0.65

aucus

-0.64

SAR

-0.64

ndum

-0.63

POSITIVE LOGITS

ually

1.16

 habits

1.12

uated

1.04

uation

1.03

 habit

0.93

uate

0.86

hered

0.74

uating

0.72

ality

0.72

iliar

0.72

Activations Density 0.046%

the word "habit" or phrases related to habits

references to behavioral habits

No Comments

No Known Activations