INDEX

Explanations

legal or official documents and statements containing negative information

oai_token-act-pair · gpt-3.5-turbo

terms related to legal documents or formal statements

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ãĤ¤ãĥĪ

-0.78

ntil

-0.75

ventus

-0.70

NetMessage

-0.68

cult

-0.64

ardless

-0.63

å§«

-0.63

âĹ¼

-0.63

past

-0.62

ecause

-0.62

POSITIVE LOGITS

 concludes

1.07

 adds

1.06

 implies

1.05

 continues

1.04

 reads

1.03

 suggests

1.03

 cites

1.01

 indicates

1.01

 contains

1.01

 says

1.00

Activations Density 0.158%

legal or official documents and statements containing negative information

terms related to legal documents or formal statements

No Comments

No Known Activations