INDEX

Explanations

words related to political conflict or military operations

oai_token-act-pair · gpt-3.5-turbo

references to guerrilla-related terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

hower

-0.83

 Sacrament

-0.73

baugh

-0.71

WARE

-0.67

ŃĶ

-0.67

 Cosponsors

-0.67

 Mour

-0.66

 Spectrum

-0.66

CLASSIFIED

-0.64

 Cycle

-0.64

POSITIVE LOGITS

inea

1.06

arant

0.94

arding

0.92

gu

0.90

vernment

0.89

ilts

0.89

ilt

0.89

pload

0.87

hou

0.80

anas

0.80

Activations Density 0.005%

words related to political conflict or military operations

references to guerrilla-related terminology

No Comments

No Known Activations