INDEX

Explanations

mentions of rankings or finishing positions in competitions

oai_token-act-pair · gpt-3.5-turbo

rankings or positions related to contests or competitions

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

hod

-0.73

natureconservancy

-0.67

 miscon

-0.66

ONSORED

-0.65

irlf

-0.64

kinson

-0.63

util

-0.62

yll

-0.62

oÄŁ

-0.62

Appearance

-0.61

POSITIVE LOGITS

holder

0.86

 bidder

0.82

holders

0.82

 standings

0.78

manship

0.78

ĪĴ

0.74

ieve

0.73

 contender

0.72

 Awakens

0.71

®

0.70

Activations Density 0.033%

mentions of rankings or finishing positions in competitions

rankings or positions related to contests or competitions

No Comments

No Known Activations