INDEX

Explanations

margin of victory or defeat in numerical terms and the term "by" that precedes it

oai_token-act-pair · gpt-3.5-turbo

phrases that indicate a score or margin in elections

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

iHUD

-0.72

jad

-0.69

 largeDownload

-0.68

Í

-0.66

SPONSORED

-0.62

âĨĳ

-0.62

ãĥīãĥ©

-0.61

 commenced

-0.58

FO

-0.58

hai

-0.58

POSITIVE LOGITS

 virtue

1.09

 leaps

0.97

 landslide

0.95

 unanimous

0.94

election

0.91

 sheer

0.79

 margins

0.77

 knockout

0.76

FAR

0.75

 landsl

0.74

Activations Density 0.062%

margin of victory or defeat in numerical terms and the term "by" that precedes it

phrases that indicate a score or margin in elections

No Comments

No Known Activations