INDEX

Explanations

information related to different geographical locations

oai_token-act-pair · gpt-3.5-turbo

references to behavior patterns or phenomena related to health and societal issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

schild

-0.62

tein

-0.61

apo

-0.59

lying

-0.57

IRA

-0.53

mbudsman

-0.53

DonaldTrump

-0.52

chens

-0.52

objects

-0.52

ebus

-0.51

POSITIVE LOGITS

 Robot

0.64

sci

0.64

 hilar

0.60

 Sega

0.60

 themed

0.59

 Comic

0.58

 Cartoon

0.57

 Anime

0.57

nown

0.57

isode

0.57

Activations Density 1.744%

information related to different geographical locations

references to behavior patterns or phenomena related to health and societal issues

No Comments

No Known Activations