INDEX

Explanations

terms related to compliance with regulations and standards

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

phrases that indicate compliance with regulations or standards

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 12-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.11.hook_resid_post

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.11.hook_resid_post

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

stros

-0.74

NetMessage

-0.74

isol

-0.67

ppa

-0.67

watching

-0.65

 frivol

-0.65

 secondly

-0.65

john

-0.65

cephal

-0.64

rade

-0.62

POSITIVE LOGITS

 stringent

1.05

 regulations

1.05

 applicable

1.04

 strict

1.04

 stricter

0.97

 rules

0.97

 laws

0.97

 guidelines

0.96

 deadlines

0.95

 federal

0.92

Activations Density 0.174%

terms related to compliance with regulations and standards

phrases that indicate compliance with regulations or standards

No Comments

No Known Activations