INDEX

Explanations

references to consumer-related terms and topics

oai_token-act-pair · gpt-3.5-turbo

references to consumers and their experiences or rights

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

 Thing

-0.66

tein

-0.63

Leader

-0.59

 Goat

-0.58

staff

-0.57

 Teach

-0.57

yss

-0.57

arp

-0.57

 Stall

-0.57

FLAG

-0.56

POSITIVE LOGITS

umers

1.39

ervatives

0.92

hip

0.92

 beware

0.87

cript

0.84

folk

0.76

pace

0.75

iband

0.75

umer

0.75

wana

0.74

Activations Density 0.034%

references to consumer-related terms and topics

references to consumers and their experiences or rights

No Comments

No Known Activations