INDEX

Explanations

computer software and technology-related terms or concepts

oai_token-act-pair · gpt-3.5-turbo

mentions of technical products, systems, and academic or security contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 9-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.9.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.9.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

anon

-0.73

 Cheong

-0.63

venants

-0.63

leon

-0.63

 Moff

-0.62

 Activate

-0.61

gall

-0.61

heid

-0.61

lves

-0.59

llah

-0.58

POSITIVE LOGITS

 purposes

1.86

 sake

1.78

ummies

1.06

 reasons

1.01

 purpose

0.95

 lovers

0.82

 beginners

0.73

 foreseeable

0.73

 reason

0.73

 duration

0.72

Activations Density 0.845%

computer software and technology-related terms or concepts

mentions of technical products, systems, and academic or security contexts

No Comments

No Known Activations