INDEX

Explanations

prepositions indicating proximity or location, specifically the word "next"

oai_token-act-pair · gpt-3.5-turbo

the phrase "next to" indicating spatial relationships

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

 Cosponsors

-0.79

OY

-0.75

udeb

-0.72

inet

-0.72

dayName

-0.71

ongo

-0.70

anke

-0.67

Cong

-0.67

ucci

-0.66

atis

-0.65

POSITIVE LOGITS

 hers

0.78

whelming

0.74

rium

0.68

sighted

0.68

 parchment

0.67

 theirs

0.65

 toast

0.64

shoot

0.62

pping

0.61

dry

0.61

Activations Density 0.035%

prepositions indicating proximity or location, specifically the word "next"

the phrase "next to" indicating spatial relationships

No Comments

No Known Activations