INDEX

Explanations

phrases within quotation marks

oai_token-act-pair · gpt-3.5-turbo

quotations and their accompanying contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

ilst

-0.75

cend

-0.75

ĻĤ

-0.71

ancies

-0.70

 mosqu

-0.70

stant

-0.69

omorphic

-0.69

onite

-0.68

etheless

-0.68

thood

-0.66

POSITIVE LOGITS

/"

1.29

SPONSORED

0.96

>>\

0.95

/>

0.79

 Whedon

0.71

DPR

0.68

;)

0.68

Spoiler

0.67

Wikipedia

0.67

sic

0.66

Activations Density 0.076%

phrases within quotation marks

quotations and their accompanying contexts

No Comments

No Known Activations