INDEX

Explanations

words related to specific names or places

oai_token-act-pair · gpt-3.5-turbo

proper nouns and names, particularly related to people and places

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 phantom

-0.61

 faint

-0.58

ADVERTISEMENT

-0.58

Interstitial

-0.57

 prank

-0.57

 unheard

-0.56

à¤

-0.56

Spoiler

-0.56

 unwelcome

-0.55

 undercut

-0.55

POSITIVE LOGITS

anski

0.80

itsch

0.77

chini

0.75

zag

0.75

illi

0.72

otti

0.72

heng

0.71

geist

0.69

abwe

0.68

Gi

0.68

Activations Density 0.116%

words related to specific names or places

proper nouns and names, particularly related to people and places

No Comments

No Known Activations