INDEX

Explanations

Apologies, but the provided text is too long and involves multiple intricate details. Could you kindly provide a shorter snippet or a summary so I can better understand the overall theme or pattern the neuron is detecting?

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

New Auto-Interp

Configuration

jbloom/Gemma-2b-Residual-Stream-SAEs/gemma_2b_blocks.6.hook_resid_post_16384_anthropic_fast_lr

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

torch.float32

Hook Point

blocks.6.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

<bos>

-3.04

/***

-0.82

<?

-0.75

<?

-0.73

///**

-0.69

//{

-0.69

})();

-0.67

/*!

-0.65

/*++

-0.65

POSITIVE LOGITS

 thut

1.31

 pymysql

1.30

fta

1.25

 wien

1.22

 maneu

1.21

Juf

1.21

 meis

1.19

ftu

1.18

fup

1.15

 stockholm

1.13

Activations Density 0.180%

Apologies, but the provided text is too long and involves multiple intricate details. Could you kindly provide a shorter snippet or a summary so I can better understand the overall theme or pattern the neuron is detecting?

No Comments

No Known Activations