INDEX

Explanations

the name "Nicole" occurring in texts

oai_token-act-pair · gpt-3.5-turbo

mentions of the name "Nicole."

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ansas

-0.71

izoph

-0.69

cgi

-0.66

committee

-0.66

HAHAHAHA

-0.65

ground

-0.65

direction

-0.65

Liter

-0.65

travel

-0.65

è¦ļéĨĴ

-0.65

POSITIVE LOGITS

 Nicole

0.98

oshenko

0.93

 Scher

0.85

lette

0.83

tti

0.83

 Loren

0.79

 Richie

0.76

oko

0.75

Ã¡s

0.75

 Holder

0.73

Activations Density 0.009%

the name "Nicole" occurring in texts

mentions of the name "Nicole."

No Comments

No Known Activations