INDEX

Explanations

phrases that indicate a relationship of belonging or association

oai_token-act-pair · gpt-3.5-turbo

the word "which" in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 7-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.7.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.7.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

icut

-0.65

Temp

-0.64

 Frie

-0.63

ÃŃa

-0.59

Tue

-0.59

DEBUG

-0.58

missing

-0.57

Live

-0.57

saw

-0.56

gra

-0.56

POSITIVE LOGITS

merce

0.92

 extent

0.82

izen

0.73

 detriment

0.69

soever

0.69

 credit

0.68

ordan

0.66

HL

0.63

izens

0.62

\/\/

0.62

Activations Density 0.034%

phrases that indicate a relationship of belonging or association

the word "which" in various contexts

No Comments

No Known Activations