INDEX

Explanations

references to specific characters or entities within parentheses

oai_token-act-pair · gpt-3.5-turbo

closing parentheses in the context of discussions or clarifications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 1-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.1.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.1.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

 metic

-0.71

¬¼

-0.69

asha

-0.67

 tremend

-0.66

©¶æ

-0.63

¢

-0.62

enta

-0.61

 dream

-0.61

ert

-0.60

iku

-0.60

POSITIVE LOGITS

 Flavoring

0.89

0.75

 however

0.73

then

0.73

 namely

0.72

etc

0.72

]),

0.70

 whereas

0.68

 nevertheless

0.66

 Anyway

0.65

Activations Density 0.056%

references to specific characters or entities within parentheses

closing parentheses in the context of discussions or clarifications

No Comments

No Known Activations