INDEX

Explanations

phrases related to the individual 'Edward Snowden'

oai_token-act-pair · gpt-3.5-turbo

mentions of Edward Snowden

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 2-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.2.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.2.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

ournals

-0.79

rian

-0.78

reek

-0.75

ogun

-0.74

rontal

-0.70

iliated

-0.70

amins

-0.69

hetical

-0.69

NES

-0.68

opathic

-0.67

POSITIVE LOGITS

 Snowden

1.19

DragonMagazine

0.83

 Greenwald

0.77

 whistleblowers

0.73

 Journalism

0.71

bin

0.70

 revelations

0.69

abad

0.67

 Files

0.65

 rall

0.64

Activations Density 0.010%

phrases related to the individual 'Edward Snowden'

mentions of Edward Snowden

No Comments

No Known Activations