INDEX

Explanations

programming-related terms and phrases indicating actions or states

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

PMailer

-0.45

 cenas

-0.43

 ček

-0.40

 flexibility

-0.39

 gouvern

-0.38

 conceptos

-0.38

 MonoBehaviour

-0.38

 harapan

-0.36

 BrowserModule

-0.36

 Vergleich

-0.35

POSITIVE LOGITS

OGND

0.57

DockStyle

0.56

бият

0.54

 мәкал

0.53

 <>",

0.51

 fornic

0.50

ConstraintMaker

0.50

PreferredItem

0.49

carros

0.48

Чыганаклар

0.47

Activations Density 1.101%

programming-related terms and phrases indicating actions or states

No Comments

No Known Activations