INDEX

Explanations

words related to rules, regulations, and policies

oai_token-act-pair · gpt-3.5-turbo

rules, restrictions, or regulations in various contexts

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

nesota

-0.82

atform

-0.75

bernatorial

-0.72

ibliography

-0.72

cellaneous

-0.66

 resurrection

-0.65

 circumst

-0.64

aukee

-0.64

ospels

-0.64

 Preview

-0.63

POSITIVE LOGITS

 quotas

1.09

 forcing

0.98

 forbid

0.97

 depri

0.96

 refusing

0.96

 forbidden

0.91

 discriminatory

0.91

 restricting

0.90

 ration

0.89

 blacklist

0.88

Activations Density 1.330%

words related to rules, regulations, and policies

rules, restrictions, or regulations in various contexts

No Comments

No Known Activations