INDEX

Explanations

phrases related to extremes, notably strong emotional reactions or contrasts

oai_token-act-pair · gpt-3.5-turbo

negative emotional states and expressions

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

yne

-0.70

 Persons

-0.62

 Families

-0.60

 Afric

-0.59

 Solitaire

-0.59

 Means

-0.55

 Syndrome

-0.53

 Instit

-0.53

 Transcript

-0.53

 Mississippi

-0.52

POSITIVE LOGITS

 upgr

0.73

uay

0.67

efully

0.67

borgh

0.66

iously

0.63

urous

0.63

ï¸

0.58

 ballpark

0.57

fitting

0.56

 fide

0.56

Activations Density 0.384%

phrases related to extremes, notably strong emotional reactions or contrasts

negative emotional states and expressions

No Comments

No Known Activations