INDEX

Explanations

terms related to specific names or entities, possibly related to medical or technical fields

oai_token-act-pair · gpt-3.5-turbo

words that end with specific suffixes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

Redditor

-0.54

ulkan

-0.51

FTWARE

-0.50

selves

-0.47

requires

-0.46

 tears

-0.46

axies

-0.46

 trope

-0.45

laws

-0.45

heric

-0.45

POSITIVE LOGITS

ĳå£«

0.78

§

0.62

undo

0.60

ago

0.59

\\\\\\\\

0.56

Īè

0.56

querque

0.55

reck

0.55

ault

0.55

efe

0.55

Activations Density 0.145%

terms related to specific names or entities, possibly related to medical or technical fields

words that end with specific suffixes

No Comments

No Known Activations