INDEX

Explanations

references to method calls or function invocations in code snippets

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

programming identifiers

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

 helft

-0.42

 jongen

-0.41

 uova

-0.40

 toiminta

-0.40

 tvor

-0.39

 actualizado

-0.36

ChromeDriver

-0.36

 mitad

-0.36

 oídos

-0.35

sweise

-0.34

POSITIVE LOGITS

0.81

 المعيارى

0.64

 EconPapers

0.63

 незавершена

0.63

 ویکی‌پدی

0.59

 مرئيه

0.58

 eventdata

0.58

:✨

0.57

 Roskov

0.55

NameInMap

0.52

Activations Density 0.013%

references to method calls or function invocations in code snippets

programming identifiers

No Comments

No Known Activations