INDEX

Explanations

mentions of the North Korean leader, Kim Jong Un

oai_token-act-pair · gpt-3.5-turbo

mentions of Kim Jong Un

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

rawdownloadcloneembedreportprint

-0.84

eport

-0.83

ONY

-0.74

WN

-0.73

WT

-0.72

 GOODMAN

-0.69

 discrep

-0.67

VID

-0.66

doi

-0.66

IGH

-0.66

POSITIVE LOGITS

 Jinping

0.82

 regime

0.80

 confid

0.77

Jr

0.73

 himself

0.73

 Hussein

0.73

Assad

0.71

 Himself

0.70

 Putin

0.69

tsy

0.69

Activations Density 0.117%

mentions of the North Korean leader, Kim Jong Un

mentions of Kim Jong Un

No Comments

No Known Activations