INDEX

Explanations

words or acronyms related to a specific organization or concept, which is likely "SI."

oai_token-act-pair · gpt-3.5-turbo

mentions of the publication "Sports Illustrated" and its associated content

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

woman

-0.84

notice

-0.81

night

-0.76

side

-0.76

EStream

-0.71

busters

-0.71

stroke

-0.71

giving

-0.70

guard

-0.70

places

-0.69

POSITIVE LOGITS

SI

1.21

SI

0.95

KK

0.88

EG

0.88

UF

0.87

FFER

0.86

OLOG

0.80

UM

0.79

ZI

0.78

BILITIES

0.78

Activations Density 0.004%

words or acronyms related to a specific organization or concept, which is likely "SI."

mentions of the publication "Sports Illustrated" and its associated content

No Comments

No Known Activations