INDEX

Explanations

headings or structural elements commonly found in reviews and academic papers

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

section headers or labels

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 otomatig

-0.80

 Biôgrafia

-0.75

sizeCache

-0.75

ParallelGroup

-0.74

complexContent

-0.73

 kasarigan

-0.73

rungsseite

-0.73

 Paglinawan

-0.73

 oprot

-0.72

IndentedString

-0.71

POSITIVE LOGITS

is

0.32

0.31

–

0.30

—

0.30

ôtel

0.28

 please

0.28

0.27

temp

0.27

Activations Density 0.612%

headings or structural elements commonly found in reviews and academic papers

section headers or labels

No Comments

No Known Activations