INDEX

Explanations

names of news agencies or sources within the text

oai_token-act-pair · gpt-3.5-turbo

repeated mentions of the term "Associated Press."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

ysis

-0.77

wagen

-0.68

mable

-0.66

tty

-0.66

bler

-0.66

aneers

-0.65

ilo

-0.65

ilk

-0.64

hang

-0.64

apo

-0.61

POSITIVE LOGITS

 Press

1.03

 Images

0.83

Images

0.79

ãĤ¤ãĥĪ

0.77

 Newsp

0.73

 Franch

0.73

itions

0.71

Org

0.70

 Content

0.69

 Correspond

0.68

Activations Density 0.028%

names of news agencies or sources within the text

repeated mentions of the term "Associated Press."

No Comments

No Known Activations