INDEX

Explanations

references to verbs and actions in the context of functions and processing

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

action or process verbs

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 PyTuple

-0.45

FieldNumber

-0.40

JsonInclude

-0.36

ACHEL

-0.36

IsMutable

-0.35

allowNull

-0.35

 InputDecoration

-0.35

TRIBUTES

-0.34

 apartment

-0.34

waren

-0.34

POSITIVE LOGITS

RenderAtEndOf

0.71

 '\\;'

0.62

tagHelperRunner

0.59

 estekak

0.50

цездатний

0.50

Tembelea

0.49

 Италијани

0.48

httphttps

0.48

QMetaType

0.47

posedge

0.47

Activations Density 0.411%

references to verbs and actions in the context of functions and processing

action or process verbs

No Comments

No Known Activations