INDEX

Explanations

mentions of the word "quite" followed by a positive adjective

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

New Auto-Interp

Configuration

jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.0.hook_resid_post_16384_anthropic

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

torch.float32

Hook Point

blocks.0.hook_resid_post

Architecture

standard

Context Size

1,024

Dataset

HuggingFaceFW/fineweb

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ⓧ

-1.08

<bos>

-1.06

/**

-1.03

-0.90

 "..\..\..\

-0.82

/*

-0.79

<?

-0.75

 considère

-0.67

 Betracht

-0.62

 croit

-0.62

POSITIVE LOGITS

 maroc

1.25

 ananas

1.24

 seksi

1.20

 alkoh

1.14

 silikon

1.14

 kafe

1.13

 peculi

1.12

 karton

1.12

 keramik

1.11

 sappi

1.10

Activations Density 0.219%

mentions of the word "quite" followed by a positive adjective

No Comments

No Known Activations