INDEX

Explanations

words related to the actions or states involving movement

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

potyvirid

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

 Lainnya

-0.60

 reducido

-0.52

 Egipto

-0.51

 virgen

-0.50

 aislado

-0.49

 Belanda

-0.49

 contigo

-0.49

 Contactez

-0.48

 reciente

-0.48

 Ambiental

-0.48

POSITIVE LOGITS

mybatisplus

0.63

 المعيارى

0.54

 ſch

0.49

vh

0.49

 unſ

0.49

NST

0.49

CCT

0.48

 الس

0.47

PhysRevD

0.46

IFA

0.44

Activations Density 0.172%

words related to the actions or states involving movement

potyvirid

No Comments

No Known Activations