INDEX

Explanations

statistical terms and data-related concepts

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

within code or markup languages

np_acts-logits-general · gemini-2.0-flash

visual links and drawing elements

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

zzar

-0.45

########.

-0.43

⟬

-0.40

 Experten

-0.39

 متعلقه

-0.39

 expertos

-0.37

UVWXYZ

-0.37

 mắn

-0.36

retanto

-0.36

Şi

-0.36

POSITIVE LOGITS

link

0.67

links

0.63

Хьажоргаш

0.60

AsUp

0.58

 link

0.57

 Link

0.56

Link

0.54

 LINK

0.52

 Roskov

0.52

AddTagHelper

0.51

Activations Density 0.012%

statistical terms and data-related concepts

within code or markup languages

visual links and drawing elements

No Comments

No Known Activations