INDEX

Explanations

names of specific individuals or entities

oai_token-act-pair · gpt-3.5-turbo

proper nouns related to individuals and their associations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.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.11.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

 Wilmington

-0.88

 Uncharted

-0.81

 Olympus

-0.80

 Sounders

-0.77

 Newport

-0.75

 Chilean

-0.75

 Californ

-0.75

 Normandy

-0.72

 Chloe

-0.71

 Scient

-0.71

POSITIVE LOGITS

uddin

1.72

 Hussain

1.71

ullah

1.67

 Mohammad

1.51

 Ahmed

1.46

 Mahm

1.45

 Imran

1.43

abad

1.42

 Shah

1.40

 Sharif

1.36

Activations Density 0.217%

names of specific individuals or entities

proper nouns related to individuals and their associations

No Comments

No Known Activations