INDEX

Explanations

programming constructs, specifically function declarations and control flow statements

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

Code and programming terms

np_acts-logits-general · gemini-2.0-flash

programming language constructs

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

</strong>

-0.54

</u>

-0.54

.”

-0.42

”.

-0.40

——

-0.39

dat

-0.38

”,

-0.37

↵

-0.36

<u>

-0.36

POSITIVE LOGITS

########.

1.15

 resourceCulture

1.10

請繼續往下閱讀

1.09

دانشنامهٔ

1.08

rungsseite

1.02

:✨

1.01

 يتيمه

1.00

 CreateTagHelper

0.98

PerformLayout

0.97

Portály

0.97

Activations Density 0.318%

programming constructs, specifically function declarations and control flow statements

Code and programming terms

programming language constructs

No Comments

No Known Activations