INDEX

Explanations

the name "Vladimir Putin."

oai_token-act-pair · gpt-3.5-turbo

references to Vladimir Putin

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

 includ

-0.76

mble

-0.74

dress

-0.69

orph

-0.67

pell

-0.66

rawdownloadcloneembedreportprint

-0.64

htt

-0.63

 trave

-0.61

ortion

-0.60

eat

-0.60

POSITIVE LOGITS

 Putin

1.54

 Lenin

1.18

Putin

1.09

 Jinping

1.02

 Tayyip

0.96

Zh

0.96

Ily

0.93

 Vlad

0.92

 Vladimir

0.89

Nab

0.89

Activations Density 0.032%

the name "Vladimir Putin."

references to Vladimir Putin

No Comments

No Known Activations