INDEX

Explanations

phrases related to administrative processes or procedures

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

<bos>

-1.99

/**

-1.28

/***

-1.27

ⓧ

-1.21

-1.16

<?

-1.15

<?

-1.11

///**

-1.03

/*

-0.92

//*/

-0.90

POSITIVE LOGITS

 kaos

1.00

 seksi

0.94

 keramik

0.86

 kafe

0.85

 silikon

0.84

 lele

0.84

 mikrofon

0.83

 balon

0.82

 panik

0.81

 optik

0.79

Activations Density 1.271%

phrases related to administrative processes or procedures

No Comments

No Known Activations