INDEX

Explanations

references to the hockey team "Montreal Canadiens" and related terms

oai_token-act-pair · gpt-3.5-turbo

mentions of specific teams, individuals, or keywords associated with sports and popular culture

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

acular

-0.79

orative

-0.76

 contempor

-0.75

eele

-0.74

uality

-0.70

ãĥ¯

-0.70

illary

-0.69

Ô

-0.68

ando

-0.68

oup

-0.68

POSITIVE LOGITS

mong

0.84

ttle

0.77

 Hoff

0.76

coh

0.72

IGH

0.72

irsch

0.70

hak

0.68

wives

0.68

 hiber

0.67

SOL

0.66

Activations Density 0.041%

references to the hockey team "Montreal Canadiens" and related terms

mentions of specific teams, individuals, or keywords associated with sports and popular culture

No Comments

No Known Activations