INDEX

Explanations

words related to scenarios or activities involving "textures" or "surfaces"

oai_token-act-pair · gpt-3.5-turbo

words related to cold beverages, particularly iced drinks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

stable

-0.78

\\\\\\\\

-0.78

yssey

-0.73

merce

-0.70

stage

-0.68

shr

-0.67

eatures

-0.67

gradient

-0.66

 livest

-0.66

band

-0.65

POSITIVE LOGITS

ices

0.86

heet

0.84

lli

0.80

cream

0.78

llo

0.70

erno

0.70

xual

0.69

utical

0.67

oint

0.65

lla

0.65

Activations Density 0.012%

words related to scenarios or activities involving "textures" or "surfaces"

words related to cold beverages, particularly iced drinks

No Comments

No Known Activations