INDEX

Explanations

proper nouns, particularly names

oai_token-act-pair · gpt-3.5-turbo

references to specific individuals and their characteristics

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

cknow

-0.84

estinal

-0.83

innie

-0.80

ufact

-0.78

 Lumpur

-0.71

ishers

-0.69

izontal

-0.65

pez

-0.65

archy

-0.65

quished

-0.65

POSITIVE LOGITS

erous

0.79

hurst

0.75

ylon

0.69

crow

0.67

xus

0.66

stick

0.66

son

0.62

 Stain

0.62

 Crawford

0.62

 Rica

0.61

Activations Density 0.036%

proper nouns, particularly names

references to specific individuals and their characteristics

No Comments

No Known Activations