INDEX

Explanations

terms related to standards and regulations

oai_token-act-pair · gpt-3.5-turbo

references to standards in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

tein

-0.75

 Rove

-0.72

seat

-0.70

Dak

-0.67

âĸ¬

-0.67

hunt

-0.67

hs

-0.66

cific

-0.63

sie

-0.63

rontal

-0.62

POSITIVE LOGITS

 standards

1.29

DragonMagazine

0.98

 compliant

0.95

 Standards

0.91

é¾įå¥ĳå£«

0.90

 deviations

0.88

Reviewer

0.80

 protocols

0.79

 standard

0.77

 bearer

0.75

Activations Density 0.015%

terms related to standards and regulations

references to standards in various contexts

No Comments

No Known Activations