INDEX

Explanations

proper nouns related to various topics, potentially spanning politics, sports, and entertainment

oai_token-act-pair · gpt-3.5-turbo

proper nouns, particularly names and entities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

enegger

-0.62

selves

-0.61

upt

-0.59

âĢ¦.

-0.56

ebin

-0.56

*.

-0.55

,...

-0.55

 downt

-0.55

sic

-0.55

glers

-0.54

POSITIVE LOGITS

 Profile

0.71

 condemns

0.63

ortunately

0.61

©¶æ¥µ

0.61

 âĵĺ

0.60

 hath

0.59

 metic

0.59

 acknowledges

0.57

 concludes

0.57

 specifies

0.57

Activations Density 0.484%

proper nouns related to various topics, potentially spanning politics, sports, and entertainment

proper nouns, particularly names and entities

No Comments

No Known Activations