INDEX

Explanations

proper nouns related to locations, specifically focusing on "Clara"

oai_token-act-pair · gpt-3.5-turbo

names of places or people, particularly focusing on "Clara."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 3-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.3.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.3.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

unct

-0.74

Marginal

-0.69

rm

-0.68

 initiation

-0.68

mediated

-0.67

ryu

-0.65

ebin

-0.65

ulatory

-0.65

tumblr

-0.64

ramid

-0.63

POSITIVE LOGITS

 Clara

1.16

 Oswald

0.94

iffe

0.89

asma

0.83

phia

0.82

 valley

0.74

 Valley

0.73

puter

0.72

Lens

0.71

 Lumpur

0.71

Activations Density 0.007%

proper nouns related to locations, specifically focusing on "Clara"

names of places or people, particularly focusing on "Clara."

No Comments

No Known Activations