INDEX

Explanations

numeric values related to research studies, such as years, percentages, and citation numbers

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

New Auto-Interp

Configuration

jbloom/Gemma-2b-IT-Residual-Stream-SAEs/gemma_2b_it_blocks.12.hook_resid_post_16384

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

HuggingFaceFW/fineweb

Features

16,384

Data Type

float32

Hook Name

blocks.12.hook_resid_post

Hook Layer

Architecture

standard

Context Size

1,024

Dataset

Skylion007/openwebtext

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 shenan

-1.74

 unspeak

-1.70

 McLaugh

-1.55

 hairc

-1.51

 indestru

-1.51

 gaily

-1.50

 disreg

-1.49

 cushi

-1.43

 indescri

-1.41

 McInt

-1.40

POSITIVE LOGITS

<bos>

1.03

 potest

0.71

 Walkover

0.71

 Paglinawan

0.69

 Kohlen

0.66

0.61

 elegan

0.60

0.59

♩

0.59

fatalError

0.59

Activations Density 0.038%

numeric values related to research studies, such as years, percentages, and citation numbers

No Comments

No Known Activations