INDEX

Explanations

references to measurements and specifications

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

anteroposterior radiographs

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

BufferException

-0.81

ValueStyle

-0.81

ロウィン

-0.69

 ―――――

-0.68

<unused41>

-0.65

<unused43>

-0.65

<unused68>

-0.65

<unused16>

-0.65

<unused3>

-0.65

<unused14>

-0.65

POSITIVE LOGITS

 eres

0.32

resizingMask

0.28

 proposons

0.28

mybatisplus

0.28

 кроме

0.27

 enim

0.27

webElementXpaths

0.26

estens

0.26

 beyond

0.26

ul

0.25

Activations Density 0.028%

references to measurements and specifications

anteroposterior radiographs

No Comments

No Known Activations