INDEX

Explanations

references to anonymity or anonymous actions

oai_token-act-pair · gpt-3.5-turbo

references to anonymity 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 @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.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.5.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

enegger

-0.83

NEY

-0.83

bourg

-0.79

rehens

-0.74

united

-0.74

neys

-0.73

union

-0.73

ELS

-0.73

eland

-0.72

tons

-0.70

POSITIVE LOGITS

 informant

0.97

 anonymous

0.92

onym

0.85

 whistleblower

0.84

 sourcing

0.82

 anonymity

0.82

 anonymously

0.81

ously

0.79

otine

0.75

iate

0.75

Activations Density 0.016%

references to anonymity or anonymous actions

references to anonymity in various contexts

No Comments

No Known Activations