INDEX

Explanations

references to quantities or numerical values

oai_token-act-pair · gpt-3.5-turbo

phrases that describe quantities or relationships using "of" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

 agre

-0.75

 encour

-0.69

icultural

-0.63

 streng

-0.63

esting

-0.63

 confir

-0.62

 misunder

-0.62

 sensit

-0.60

 performance

-0.59

 disapp

-0.58

POSITIVE LOGITS

teenth

0.75

teen

0.73

ĳå£«

0.69

those

0.66

eely

0.64

icial

0.63

ife

0.63

them

0.63

ridor

0.62

ARE

0.62

Activations Density 0.068%

references to quantities or numerical values

phrases that describe quantities or relationships using "of" in various contexts

No Comments

No Known Activations