INDEX

Explanations

terms related to structured projects, tools, and methodologies in various contexts

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

research and project terms

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

:+:

-0.47

 proper

-0.43

 Proper

-0.40

 ivelany

-0.39

OK

-0.39

wapV

-0.38

gettext

-0.38

bidi

-0.38

 Administrativna

-0.37

拗

-0.36

POSITIVE LOGITS

ähteet

0.60

 surla

0.57

 Infór

0.57

 miniaturka

0.54

 vocês

0.54

EDEFAULT

0.54

 consortium

0.53

 suaminya

0.52

pushFollow

0.52

 celebridades

0.52

Activations Density 0.124%

terms related to structured projects, tools, and methodologies in various contexts

research and project terms

No Comments

No Known Activations