INDEX

Explanations

proper nouns related to a specific movie or a series

oai_token-act-pair · gpt-3.5-turbo

references to specific franchises or series titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 0-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.0.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.0.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 repro

-0.73

study

-0.70

bush

-0.63

 Barn

-0.62

 advis

-0.61

 Boone

-0.60

 Conservation

-0.60

protection

-0.60

Construction

-0.59

 encour

-0.59

POSITIVE LOGITS

Ã¡s

0.83

entious

0.79

ãĥ³ãĤ¸

0.75

puter

0.73

egal

0.72

ÑĮ

0.69

urities

0.67

inho

0.67

Ã³n

0.67

nw

0.67

Activations Density 0.000%

proper nouns related to a specific movie or a series

references to specific franchises or series titles

No Comments

No Known Activations