INDEX

Explanations

the word "Janet"

oai_token-act-pair · gpt-3.5-turbo

the name "Janet" and related references in the context of authoritative figures

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

Interstitial

-0.92

-0.79

FTWARE

-0.76

juven

-0.73

doors

-0.72

gran

-0.68

Sov

-0.67

ãĥı

-0.66

ebook

-0.65

bid

-0.65

POSITIVE LOGITS

 Janet

1.19

 Mock

0.90

 Nicol

0.79

 DeVos

0.78

 Reno

0.77

ovic

0.76

jen

0.75

 Walton

0.73

ellen

0.72

 Porter

0.71

Activations Density 0.007%

the word "Janet"

the name "Janet" and related references in the context of authoritative figures

No Comments

No Known Activations