INDEX

Explanations

numerical values and symbols indicative of coding or technical contexts

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

technical document codes

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

PreferredItem

-0.61

fromnode

-0.51

WithIOException

-0.49

PerformLayout

-0.48

transQ

-0.46

 Sinf

-0.45

 Battista

-0.44

rawDesc

-0.44

 propOrder

-0.42

IntoConstraints

-0.42

POSITIVE LOGITS

modelBuilder

0.56

 мәкал

0.47

 Comprometido

0.45

 redacción

0.44

 animación

0.42

Personensuche

0.41

 חיצוניים

0.41

الإنجليزية

0.39

 rédaction

0.37

 سكانية

0.37

Activations Density 0.025%

numerical values and symbols indicative of coding or technical contexts

technical document codes

No Comments

No Known Activations