INDEX

Explanations

programming-related syntax, particularly method declarations and documentation

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

method definitions and calls

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

 Мексичка

-0.73

 setEmail

-0.63

 surla

-0.60

IsMutable

-0.60

 صوتيه

-0.58

 postIndex

-0.57

Jeografia

-0.56

 setAddress

-0.56

WithIOException

-0.55

tanleria

-0.54

POSITIVE LOGITS

fromnode

0.40

 gomma

0.37

RenderAtEndOf

0.34

Архівовано

0.33

tadora

0.33

 orientación

0.32

 Handeln

0.32

appcompat

0.32

توضیحات

0.31

yym

0.31

Activations Density 0.006%

programming-related syntax, particularly method declarations and documentation

method definitions and calls

No Comments

No Known Activations