INDEX

Explanations

references to code, programming syntax, and technical terminology

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

programming context

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

 Nowadays

-0.50

 nowadays

-0.46

ChildScrollView

-0.44

Nowadays

-0.43

 الوطنيه

-0.41

+#+#

-0.41

 mencion

-0.40

anmoins

-0.38

 recommandations

-0.37

 convain

-0.36

POSITIVE LOGITS

AndEndTag

0.62

 CreateTagHelper

0.52

Lit

0.52

 otomatig

0.52

ennifer

0.52

 Monfieur

0.51

 HIPAA

0.50

jsxFileName

0.49

 Quan

0.49

teman

0.49

Activations Density 0.059%

references to code, programming syntax, and technical terminology

programming context

No Comments

No Known Activations