INDEX

Explanations

topics or categories mentioned in a text or document

oai_token-act-pair · gpt-3.5-turbo

topic headers and keywords within a document

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

ĸļ

-0.77

Â¯

-0.77

IELD

-0.74

 subsequ

-0.73

erred

-0.71

rites

-0.69

athing

-0.69

ornia

-0.68

rier

-0.67

ards

-0.67

POSITIVE LOGITS

 Mandatory

0.66

 Become

0.64

 Various

0.61

?]

0.59

 POLIT

0.59

 Provided

0.57

 Cosponsors

0.57

econom

0.56

govtrack

0.56

education

0.56

Activations Density 0.035%

topics or categories mentioned in a text or document

topic headers and keywords within a document

No Comments

No Known Activations