INDEX

Explanations

elements related to legal and procedural contexts

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

Non-English text

np_acts-logits-general · gemini-2.0-flash

Slavic and Spanish noun endings

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_31/width_16k/average_l0_76

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Brito

-0.51

 Barrera

-0.41

 Marín

-0.39

BRE

-0.38

++++++++++++++++

-0.37

bre

-0.36

 Inoue

-0.35

udesta

-0.35

μένη

-0.33

 Pittman

-0.33

POSITIVE LOGITS

 noDo

0.69

 ब्रेकडाउन

0.66

 ſch

0.66

 linkovi

0.64

EndGlobalSection

0.64

 pleaſure

0.64

 ſche

0.62

CloseOperation

0.61

ating

0.60

 ujednoznacz

0.59

Activations Density 0.057%

elements related to legal and procedural contexts

Non-English text

Slavic and Spanish noun endings

No Comments

No Known Activations