INDEX

Explanations

phrases related to legal matters and accusations

oai_token-act-pair · gpt-3.5-turbo

references to political controversy or conflict

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 10-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.10.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.10.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

 Flavoring

-0.76

 Alpine

-0.72

 clustered

-0.70

 Twins

-0.67

 seasonal

-0.64

 Renault

-0.63

 rainy

-0.63

 Bened

-0.63

 upbeat

-0.62

LU

-0.61

POSITIVE LOGITS

 disgrace

1.29

 slander

1.19

 treason

1.17

 disrespect

1.17

 misrepresent

1.14

 hypocrisy

1.13

 unconstitutional

1.07

 dishon

1.06

 disrespectful

1.04

 insulting

1.03

Activations Density 0.858%

phrases related to legal matters and accusations

references to political controversy or conflict

No Comments

No Known Activations