INDEX

Explanations

names of individuals or locations with high relevance

oai_token-act-pair · gpt-3.5-turbo

proper nouns or names of individuals

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 envy

-0.90

FTWARE

-0.73

 ambassadors

-0.69

ModLoader

-0.68

 constants

-0.64

 sophistic

-0.63

Reviewer

-0.62

CONT

-0.60

 shorthand

-0.59

merce

-0.59

POSITIVE LOGITS

eman

0.94

uer

0.90

chuk

0.90

enberg

0.90

orst

0.89

stad

0.89

Jr

0.89

cker

0.89

kamp

0.88

linger

0.88

Activations Density 0.413%

names of individuals or locations with high relevance

proper nouns or names of individuals

No Comments

No Known Activations