INDEX

Explanations

names of individuals related to the film industry

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals and sports-related terms

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

sworth

-0.96

scl

-0.79

acca

-0.78

bnb

-0.72

undle

-0.72

anguage

-0.70

stract

-0.70

nown

-0.70

ðĿ

-0.69

ó

-0.68

POSITIVE LOGITS

 rand

0.88

 capacitor

0.72

 Renault

0.69

reet

0.68

 armour

0.68

arget

0.68

 angles

0.67

 lieu

0.67

etr

0.67

 bras

0.67

Activations Density 0.020%

names of individuals related to the film industry

references to specific individuals and sports-related terms

No Comments

No Known Activations