INDEX

Explanations

proper nouns or names related to brands, companies, or products

oai_token-act-pair · gpt-3.5-turbo

the term "original" 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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

aiman

-0.76

wal

-0.74

rooms

-0.72

ucket

-0.71

attled

-0.70

uls

-0.68

washer

-0.68

raq

-0.68

avers

-0.68

ctrl

-0.66

POSITIVE LOGITS

 Flavoring

0.96

 Original

0.91

 Poster

0.75

 Version

0.74

ity

0.74

Original

0.73

Orig

0.71

 Creator

0.69

 Score

0.69

 Trailer

0.69

Activations Density 0.013%

proper nouns or names related to brands, companies, or products

the term "original" in various contexts

No Comments

No Known Activations