INDEX

Explanations

references to possessive pronouns

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

possessive pronouns and nouns

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

 >=",

-0.46

 للاسماء

-0.44

Πηγές

-0.43

Бележки

-0.43

 виправивши

-0.42

เลย

-0.41

 raccont

-0.41

">//

-0.41

 plastica

-0.40

出版年

-0.39

POSITIVE LOGITS

 Paglinawan

0.56

migrationBuilder

0.44

/**

0.42

 favorite

0.41

own

0.38

 favourite

0.38

 CYTHON

0.37

 little

0.37

DebuggerStep

0.36

Activations Density 0.020%

references to possessive pronouns

possessive pronouns and nouns

No Comments

No Known Activations