INDEX

Explanations

phrases referring to uncertainty or lack of knowledge

oai_token-act-pair · gpt-3.5-turbo

phrases expressing uncertainty or lack of knowledge about situations or concepts

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

 Alive

-0.70

rontal

-0.69

Lago

-0.69

hement

-0.69

areth

-0.68

geoning

-0.68

efeated

-0.65

zar

-0.65

oak

-0.64

stract

-0.64

POSITIVE LOGITS

 entails

0.78

 fuss

0.75

#$

0.74

really

0.74

ãĤ´

0.70

 wrought

0.70

?",

0.70

 entail

0.70

 REALLY

0.69

 meant

0.68

Activations Density 0.396%

phrases referring to uncertainty or lack of knowledge

phrases expressing uncertainty or lack of knowledge about situations or concepts

No Comments

No Known Activations