INDEX

Explanations

references to the country of Iran and its governmental system

oai_token-act-pair · gpt-3.5-turbo

references to the Iranian government and its associated entities

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

lished

-0.88

illard

-0.88

leground

-0.86

chat

-0.86

VEL

-0.84

fill

-0.82

cial

-0.82

erk

-0.82

shaw

-0.81

iatrics

-0.81

POSITIVE LOGITS

 consulate

0.83

 dissidents

0.83

istani

0.82

 Revolutionary

0.81

 regime

0.78

 embassy

0.78

 influence

0.75

 hegemony

0.73

 exile

0.73

oil

0.72

Activations Density 0.008%

references to the country of Iran and its governmental system

references to the Iranian government and its associated entities

No Comments

No Known Activations