INDEX

Explanations

proper nouns or names, specifically the name "Oliver" with varying intensities

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Oliver"

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

mble

-0.76

ulhu

-0.71

yip

-0.70

habi

-0.70

arching

-0.69

hz

-0.69

BOOK

-0.68

evil

-0.67

ranking

-0.66

ept

-0.66

POSITIVE LOGITS

 Twist

0.90

 Oliver

0.86

sson

0.83

 Wend

0.78

alia

0.71

nel

0.70

 Crom

0.70

ator

0.68

icity

0.67

ogens

0.67

Activations Density 0.015%

proper nouns or names, specifically the name "Oliver" with varying intensities

mentions of the name "Oliver"

No Comments

No Known Activations