INDEX

Explanations

words related to behavior, specifically polite and reserved actions

oai_token-act-pair · gpt-3.5-turbo

references to social behavior and manners

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

 HUGE

-0.83

!!!!!

-0.81

!!!!

-0.79

initely

-0.78

 UPDATE

-0.77

 START

-0.75

ANA

-0.72

 TRUMP

-0.72

!!!

-0.71

 Lets

-0.71

POSITIVE LOGITS

 scorn

1.00

 indign

0.93

 unworthy

0.91

 indisc

0.90

 betray

0.88

 fleeting

0.85

 arist

0.84

 treacher

0.83

 griev

0.82

 disappro

0.82

Activations Density 1.416%

words related to behavior, specifically polite and reserved actions

references to social behavior and manners

No Comments

No Known Activations