INDEX

Explanations

references to programming constructs, specifically data structures and their serialization/deserialization processes

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

Code related to reading/writing strings and documents

np_acts-logits-general · gemini-2.0-flash

code delimiters and language syntax

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 18-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_18/width_16k/average_l0_74

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.18.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

]")]

-0.78

LayoutStyle

-0.72

 MainAxisSize

-0.71

 &___

-0.70

apimachinery

-0.70

rawDesc

-0.68

Diweddarwch

-0.66

HideFlags

-0.64

 AspNetCore

-0.64

########.

-0.63

POSITIVE LOGITS

 Sosten

0.50

')(

0.47

 fér

0.45

 nhat

0.44

 CURIAM

0.43

編

0.42

(()

0.41

 Tanto

0.41

StartTag

0.40

‪

0.40

Activations Density 0.679%

references to programming constructs, specifically data structures and their serialization/deserialization processes

Code related to reading/writing strings and documents

code delimiters and language syntax

No Comments

No Known Activations