INDEX

Explanations

phrases indicating reluctance or refusal to disclose information

oai_token-act-pair · gpt-3.5-turbo

instances of refusal or avoidance to comment or provide information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 11-res-jb

Configuration

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

soDeliveryDate

-0.96

 Nurs

-0.72

pires

-0.71

asonic

-0.68

 Compass

-0.64

govtrack

-0.61

quarters

-0.61

Veh

-0.60

 Millennium

-0.60

selves

-0.60

POSITIVE LOGITS

 comment

1.56

 speculate

1.55

 disclose

1.47

 elaborate

1.41

 specify

1.34

 divul

1.33

 discuss

1.28

 characterize

1.26

 reveal

1.22

 clarify

1.20

Activations Density 0.087%

phrases indicating reluctance or refusal to disclose information

instances of refusal or avoidance to comment or provide information

No Comments

No Known Activations