INDEX

Explanations

commonalities or repetitions in a given context, such as repeated phrases, locations, or scenarios

oai_token-act-pair · gpt-3.5-turbo

concepts and items associated with sameness and repetition

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 8-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.8.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.8.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

unin

-0.56

chieve

-0.54

 embr

-0.54

 bystand

-0.53

inker

-0.52

Emer

-0.52

 STAR

-0.51

HEL

-0.51

particularly

-0.51

 Marshal

-0.51

POSITIVE LOGITS

same

0.90

kefeller

0.74

 twice

0.72

 applies

0.67

nings

0.67

natureconservancy

0.67

ietal

0.64

 principals

0.63

igne

0.61

ait

0.60

Activations Density 0.173%

commonalities or repetitions in a given context, such as repeated phrases, locations, or scenarios

concepts and items associated with sameness and repetition

No Comments

No Known Activations