INDEX

Explanations

terms related to data normalization and conditioning processes

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

normalization and renormalization

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_24/width_131k/average_l0_97

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.24.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

########.

-0.57

itemBuilder

-0.49

 rurales

-0.48

一头

-0.47

 edible

-0.47

tagext

-0.47

 Edible

-0.47

loz

-0.47

skraft

-0.47

 amet

-0.46

POSITIVE LOGITS

BatchNorm

1.31

ulemon

0.39

 Diener

0.37

 InputDecoration

0.37

norm

0.36

SafeMath

0.35

 regulate

0.34

EnableWeb

0.34

Demografie

0.33

 winter

0.33

Activations Density 0.001%

terms related to data normalization and conditioning processes

normalization and renormalization

No Comments

No Known Activations