INDEX

Explanations

specific variable names or identifiers used in programming-related contexts

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

hex codes and identifiers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 31-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_131k/average_l0_109

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.31.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

addGap

-0.55

testify

-0.46

 onPostExecute

-0.45

 noDo

-0.43

EditorBrowsable

-0.43

Ono

-0.43

ﾟ)

-0.42

BindingResult

-0.42

opedia

-0.42

ecake

-0.41

POSITIVE LOGITS

 navideños

0.57

alakip

0.56

Demografie

0.52

 tránsito

0.49

 Grüsse

0.48

 navideñas

0.47

 bonté

0.47

 navideña

0.45

uestamente

0.45

 trânsito

0.44

Activations Density 0.001%

specific variable names or identifiers used in programming-related contexts

hex codes and identifiers

No Comments

No Known Activations