INDEX

Explanations

mentions of a specific media outlet, "The Washington Post," within text

oai_token-act-pair · gpt-3.5-turbo

repeated mentions of "The Washington Post."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

sed

-0.84

wcsstore

-0.78

oaded

-0.77

umpy

-0.77

utory

-0.74

cled

-0.74

ãĤ¤ãĥĪ

-0.73

tesy

-0.72

culated

-0.72

inding

-0.72

POSITIVE LOGITS

 Tribune

0.86

greSQL

0.84

 Dispatch

0.84

 Office

0.84

 Courier

0.82

 Magazine

0.81

 Herald

0.80

 Gazette

0.80

gres

0.79

 Staff

0.77

Activations Density 0.014%

mentions of a specific media outlet, "The Washington Post," within text

repeated mentions of "The Washington Post."

No Comments

No Known Activations