INDEX

Explanations

adjectives and verbs expressing compatibility or synchronicity

oai_token-act-pair · gpt-3.5-turbo

concepts of compatibility and interconnectedness

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

bern

-0.73

unc

-0.69

phies

-0.67

aina

-0.64

agara

-0.61

ighth

-0.60

anche

-0.59

ĳå£«

-0.59

erest

-0.59

guyen

-0.58

POSITIVE LOGITS

 twins

1.20

 between

1.13

 oppos

1.05

 twin

0.98

 Between

0.91

 sexes

0.91

 with

0.91

between

0.91

icut

0.90

 geographically

0.88

Activations Density 0.152%

adjectives and verbs expressing compatibility or synchronicity

concepts of compatibility and interconnectedness

No Comments

No Known Activations