INDEX

Explanations

words related to physical structures, specifically walls

oai_token-act-pair · gpt-3.5-turbo

references to physical barriers or structural components

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

Published

-0.72

Gene

-0.71

avez

-0.69

heny

-0.68

ya

-0.66

Wild

-0.65

phrine

-0.65

EVA

-0.65

Pacific

-0.65

 PubMed

-0.64

POSITIVE LOGITS

 walls

1.31

papers

1.00

 Walls

0.97

 wall

0.94

aby

0.90

abies

0.88

 ceilings

0.78

paper

0.78

 paintings

0.77

eries

0.76

Activations Density 0.011%

words related to physical structures, specifically walls

references to physical barriers or structural components

No Comments

No Known Activations