INDEX

Explanations

phrases related to safety instructions and precautions

oai_token-act-pair · gpt-3.5-turbo

expressions related to safety and precautionary measures

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

 Krugman

-0.88

 economists

-0.78

 Economist

-0.75

galitarian

-0.75

 invented

-0.72

 narrator

-0.71

 Gingrich

-0.70

 Nobel

-0.69

 Gins

-0.69

 Democr

-0.69

POSITIVE LOGITS

 disturbance

1.01

 incidents

0.96

angering

0.95

 impacted

0.94

 incident

0.91

 personnel

0.91

 offences

0.90

 hazardous

0.89

 offenders

0.88

 Incident

0.88

Activations Density 0.774%

phrases related to safety instructions and precautions

expressions related to safety and precautionary measures

No Comments

No Known Activations