INDEX

Explanations

identifiers and data structure components related to programming or software development

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

identifier assignments

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

featureID

-1.05

 ModelExpression

-0.99

WriteTagHelper

-0.97

 autorytatywna

-0.96

InjectAttribute

-0.95

 CreateTagHelper

-0.94

 ब्रेकडाउन

-0.92

 cherchés

-0.85

DeleteBehavior

-0.82

 Administrativna

-0.82

POSITIVE LOGITS

0.47

0.38

0.35

 result

0.34

0.33

 that

0.33

Activations Density 0.084%

identifiers and data structure components related to programming or software development

identifier assignments

No Comments

No Known Activations