INDEX

Explanations

occurrences of the term "class" in various contexts

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

user prompts

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

Jîn

-0.59

addCriterion

-0.57

 Wikimedijinoj

-0.50

XMLSchema

-0.44

 tiguan

-0.44

TRAILING

-0.44

KommentareTeilen

-0.44

httphttps

-0.43

LEADING

-0.42

addGap

-0.42

POSITIVE LOGITS

 MainActivity

0.44

+#+

0.42

argout

0.41

 Mascul

0.36

 ViewController

0.36

 Happiness

0.35

 ***!

0.35

Coarse

0.34

fitri

0.34

 Travelers

0.33

Activations Density 0.000%

occurrences of the term "class" in various contexts

user prompts

No Comments

No Known Activations