INDEX

Explanations

elements and attributes related to programming and data structures

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

Mathematical or programming notation

np_acts-logits-general · gemini-2.0-flash

the class

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

stdc

-0.33

 >=",

-0.28

 kaarangay

-0.27

).__

-0.27

 izay

-0.25

-0.23

 Paglinawan

-0.23

 Wikimedijinoj

-0.23

EXTERN

-0.23

 ویکی‌پدی

-0.23

POSITIVE LOGITS

 fashiola

0.68



0.68



0.67

 queſto

0.66

 tartalo

0.66

쎲

0.65



0.65

<unused36>

0.65

<unused20>

0.64

<unused14>

0.64

Activations Density 1.953%

elements and attributes related to programming and data structures

Mathematical or programming notation

the class

No Comments

No Known Activations