INDEX

Explanations

file formats and related metadata

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

file names with extensions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.20.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

حياته

-0.66

 AssemblyCulture

-0.46

ftra

-0.46

ParallelGroup

-0.45

 freeing

-0.44

 puissent

-0.44

حياتها

-0.43

+:+

-0.43

urably

-0.41

 kasarigan

-0.41

POSITIVE LOGITS

indd

0.70

InjectAttribute

0.56

 referrerpolicy

0.50

 defaultstate

0.44

richTextPanel

0.43

 downloaded

0.42

RegistryLite

0.41

 zipfile

0.40

jpg

0.40

SerializedName

0.39

Activations Density 0.078%

file formats and related metadata

file names with extensions

No Comments

No Known Activations