INDEX

Explanations

high-ranking political figures and officials

oai_token-act-pair · gpt-3.5-turbo

titles and positions of political leaders and officials

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

atum

-0.65

Sensor

-0.64

Limited

-0.61

ikawa

-0.61

Tuesday

-0.60

Christmas

-0.60

estival

-0.59

Market

-0.59

Tree

-0.58

Request

-0.57

POSITIVE LOGITS

 Saddam

0.88

 disgr

0.83

 Lyndon

0.80

 Sheila

0.79

turned

0.79

 Newt

0.79

 Gaddafi

0.75

 Barack

0.75

 Rick

0.75

Jim

0.74

Activations Density 0.121%

high-ranking political figures and officials

titles and positions of political leaders and officials

No Comments

No Known Activations