INDEX

Explanations

numerical or statistical data

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

grant/case/patent citations

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

rungsseite

-0.73

WriteTagHelper

-0.69

 dezelve

-0.67

 zoude

-0.66

 BoxDecoration

-0.65

 themſelves

-0.64

 juſ

-0.63

 ſich

-0.63

 zelve

-0.63

 ſein

-0.63

POSITIVE LOGITS

///</

0.43

期刊论文

0.41

 pinulongan

0.38

hdashline

0.35

 frigo

0.35

ViewImports

0.35

【

0.33

 folder

0.32

 Activity

0.31

Activations Density 0.050%

numerical or statistical data

grant/case/patent citations

No Comments

No Known Activations