INDEX

Explanations

specific categorizations and classifications of data or subjects

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

grades and categories

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

 jouet

-0.40

 Италијани

-0.39

toy

-0.38

parsedMessage

-0.38

contentLoaded

-0.38

visející

-0.37

CodedInputStream

-0.36

WebElementEntity

-0.36

сылкі

-0.35

httphttps

-0.35

POSITIVE LOGITS

$_['

0.49

 autorytatywna

0.48

 lower

0.45

 AttributeSet

0.44

 grade

0.44

 level

0.44

 Upper

0.44

InSection

0.44

 grades

0.44

 Grade

0.43

Activations Density 0.057%

specific categorizations and classifications of data or subjects

grades and categories

No Comments

No Known Activations