INDEX

Explanations

words related to tightening or loosening regulations, controls, or restrictions

oai_token-act-pair · gpt-3.5-turbo

terms related to regulation and restrictions on policy

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

 Bucc

-0.76

Ì¶

-0.70

 Finished

-0.70

 MISS

-0.69

plain

-0.66

Amazing

-0.66

 Illusion

-0.64

false

-0.64

 untold

-0.63

 Clever

-0.62

POSITIVE LOGITS

 penalties

1.34

 restrictions

1.32

 punishments

1.27

 enforcement

1.26

 regulations

1.26

 sanctions

1.22

 crackdown

1.17

 regulation

1.09

 rules

1.08

 protections

1.08

Activations Density 0.249%

words related to tightening or loosening regulations, controls, or restrictions

terms related to regulation and restrictions on policy

No Comments

No Known Activations