INDEX

Explanations

the names of political figures and positions

oai_token-act-pair · gpt-3.5-turbo

references to former 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 @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

verse

-0.62

Located

-0.61

EngineDebug

-0.61

Christmas

-0.59

 Accessed

-0.58

Tuesday

-0.58

Monday

-0.57

 endpoint

-0.57

Limited

-0.56

 fingert

-0.56

POSITIVE LOGITS

 Saddam

0.86

 Rudy

0.84

 Fidel

0.84

 disgr

0.83

 Newt

0.83

 Boris

0.78

 Gore

0.77

 Mikhail

0.74

 Ferdinand

0.74

 Alan

0.73

Activations Density 0.123%

the names of political figures and positions

references to former political leaders and officials

No Comments

No Known Activations