INDEX

Explanations

proper nouns or names, specifically "Ken" with varying suffixes

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Ken."

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.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.4.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

xual

-0.88

anwhile

-0.83

 confir

-0.79

 exha

-0.77

Reloaded

-0.71

pmwiki

-0.70

 sensitive

-0.69

ylum

-0.66

 plural

-0.66

itous

-0.66

POSITIVE LOGITS

yon

1.00

ji

0.94

ois

0.94

zie

0.94

wei

0.87

ichi

0.87

zo

0.84

 Mans

0.82

zen

0.82

Ken

0.81

Activations Density 0.008%

proper nouns or names, specifically "Ken" with varying suffixes

mentions of the name "Ken."

No Comments

No Known Activations