INDEX

Explanations

statements or instances of apologies

oai_token-act-pair · gpt-3.5-turbo

occurrences of the word "apology" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

everal

-0.84

eways

-0.78

arnaev

-0.75

eteen

-0.73

ynski

-0.72

laws

-0.71

ravis

-0.70

roups

-0.65

odiac

-0.64

alez

-0.64

POSITIVE LOGITS

 apology

0.84

cffffcc

0.81

":""},{"

0.72

xual

0.72

 announcement

0.71

AMI

0.71

uracy

0.71

ishment

0.71

 CLSID

0.70

istle

0.70

Activations Density 0.016%

statements or instances of apologies

occurrences of the word "apology" in various contexts

No Comments

No Known Activations