INDEX

Explanations

words related to physical actions or movements

oai_token-act-pair · gpt-3.5-turbo

words related to items, classifications, or categories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

omez

-0.67

ÄŁ

-0.66

DO

-0.65

andise

-0.63

Bok

-0.62

Published

-0.59

ADA

-0.57

Guest

-0.57

ATTLE

-0.56

Oz

-0.56

POSITIVE LOGITS

haps

0.79

 nipples

0.79

rums

0.72

Cola

0.68

pling

0.67

oming

0.65

ographed

0.65

bilt

0.65

adium

0.64

umatic

0.63

Activations Density 0.063%

words related to physical actions or movements

words related to items, classifications, or categories

No Comments

No Known Activations