INDEX

Explanations

phrases about specific topics or entities, such as product names, song lyrics, or personal names

oai_token-act-pair · gpt-3.5-turbo

references to culture and societal themes

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

 paperback

-0.75

¥µ

-0.72

 Flame

-0.69

 Boone

-0.68

 charm

-0.67

 Deadly

-0.65

 Marketable

-0.64

 flash

-0.64

 Doomsday

-0.63

 softer

-0.63

POSITIVE LOGITS

âĢ

1.88

Â

1.17

1.15

 âĢ

1.14

âĢł

1.12

âī

1.11

**

1.10

âĪ

1.09

Â´

1.08

âĶ

1.08

Activations Density 0.375%

phrases about specific topics or entities, such as product names, song lyrics, or personal names

references to culture and societal themes

No Comments

No Known Activations