INDEX

Explanations

words related to a specific name, "Ange"

oai_token-act-pair · gpt-3.5-turbo

occurrences of a specific character name 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

 pioneering

-0.70

 premium

-0.70

 stub

-0.66

 short

-0.66

Kit

-0.62

 resource

-0.62

 fueling

-0.61

driving

-0.61

senal

-0.61

 competitive

-0.61

POSITIVE LOGITS

ange

1.33

 Amph

0.95

lihood

0.85

anges

0.83

 Warfare

0.83

velt

0.81

aux

0.80

quist

0.78

izon

0.78

vernment

0.78

Activations Density 0.006%

words related to a specific name, "Ange"

occurrences of a specific character name in various contexts

No Comments

No Known Activations