INDEX

Explanations

closing brackets and parentheses in code-like text

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

Closing brackets

np_acts-logits-general · gemini-2.0-flash

closing brackets and operators

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

цездатний

-0.63

 esternos

-0.63

 AssemblyCompany

-0.59

 resourceCulture

-0.58

 kasarigan

-0.56

 ब्रेकडाउन

-0.56

 noDo

-0.54

 незавершена

-0.54

/*

-0.50

脚注の使い方

-0.48

POSITIVE LOGITS

0.42

adalajara

0.41

 Maui

0.40

 Janeiro

0.39

 betrekking

0.39

sont

0.38

Css

0.38

onomía

0.37

 paulista

0.37

 praktijk

0.36

Activations Density 0.844%

closing brackets and parentheses in code-like text

Closing brackets

closing brackets and operators

No Comments

No Known Activations