INDEX

Explanations

occurrences of code or programming-related commands and structures

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

programming functions and method calls related to writing data

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

font-variant:small-caps;

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-65k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_65k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

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

帖最后由

-0.80

Personendaten

-0.79

CopyWith

-0.79

thâu

-0.78

ContentAlignment

-0.77

bewerken

-0.77

 تضيفلها

-0.75

 igång

-0.75

KURZBESCHREIBUNG

-0.74

PreferredItem

-0.73

POSITIVE LOGITS

"</

0.51

 greater

0.47

大

0.46

la

0.46

clase

0.43

cula

0.43

La

0.42

Si

0.41

du

0.40

 createState

0.40

Activations Density 0.010%

occurrences of code or programming-related commands and structures

programming functions and method calls related to writing data

font-variant:small-caps;

No Comments

No Known Activations