INDEX

Explanations

references to information provided by anonymous sources

oai_token-act-pair · gpt-3.5-turbo

references to anonymity and official sources in discussions

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

http

-0.71

netflix

-0.70

HD

-0.66

 congr

-0.62

oplan

-0.60

ocre

-0.60

edia

-0.59

 Conquer

-0.59

 Glob

-0.59

Figure

-0.58

POSITIVE LOGITS

 anonymity

1.41

 confidentiality

1.21

 confidential

1.00

 sensitivity

0.95

 privacy

0.94

 secrecy

0.92

 nond

0.85

 candid

0.85

 condition

0.81

 sensitive

0.80

Activations Density 0.097%

references to information provided by anonymous sources

references to anonymity and official sources in discussions

No Comments

No Known Activations