INDEX

Explanations

tokens related to programming and technical commands

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

code development context

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

хьтан

-0.72

ientras

-0.63

ConstraintMaker

-0.61

ロウィン

-0.60

exitRule

-0.60

iſen

-0.60

 DriverManager

-0.58

فایل‌لار

-0.58

 NDEBUG

-0.58

lgari

-0.56

POSITIVE LOGITS

setVerticalGroup

0.43

:✨

0.40

Coming

0.37

↵

0.37

agod

0.36

 Coming

0.36

<code>

0.36

 EconPapers

0.35

WriteLiteral

0.35

writerow

0.35

Activations Density 0.006%

tokens related to programming and technical commands

code development context

No Comments

No Known Activations