INDEX

Explanations

code comments and method definitions in programming

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

programming language keywords and symbols

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

CUSSION

-0.31

</strong>

-0.28

 okay

-0.28

 trừ

-0.28

spec

-0.27

ok

-0.27

也不

-0.27

Cab

-0.26

 Autor

-0.26

 σή

-0.26

POSITIVE LOGITS

 للاسماء

0.87

richTextPanel

0.73

MigrationBuilder

0.68

 myſelf

0.66

JspWriter

0.66

IUrlHelper

0.66

elemField

0.65

RTSC

0.64

 kasarigan

0.64

setVerticalGroup

0.64

Activations Density 0.023%

code comments and method definitions in programming

programming language keywords and symbols

No Comments

No Known Activations