INDEX

Explanations

words related to investigations, inquiries, or probing into issues

oai_token-act-pair · gpt-3.5-turbo

prepositions and certain connecting words in a sentence structure

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

ilaterally

-0.74

emale

-0.65

 Seym

-0.61

dn

-0.60

 marqu

-0.60

 Clarkson

-0.58

 corrid

-0.58

FFFF

-0.57

 vulner

-0.57

amorph

-0.56

POSITIVE LOGITS

actionDate

0.89

 SHARES

0.70

 largeDownload

0.66

 UPDATE

0.65

 Flavoring

0.62

Psy

0.61

 UNIVERS

0.61

UTERS

0.61

 Steam

0.59

 Wrestling

0.58

Activations Density 0.164%

words related to investigations, inquiries, or probing into issues

prepositions and certain connecting words in a sentence structure

No Comments

No Known Activations