INDEX

Explanations

code-related structures and function definitions dealing with information properties and class behaviors

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

set

-0.56

-0.52

re

-0.50

let

-0.48

-0.47

خول

-0.46

den

-0.45

be

-0.45

en

-0.44

-0.43

POSITIVE LOGITS

 Мексичка

0.99

+#+#

0.97

 wikipagina

0.91

 AssemblyCulture

0.88

itemBuilder

0.88

 myſelf

0.87

Datuak

0.85

/**

0.85

ReusableCell

0.85

setVerticalGroup

0.84

Activations Density 0.304%

code-related structures and function definitions dealing with information properties and class behaviors

No Comments

No Known Activations