INDEX

Explanations

elements related to API versioning and specifications

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

version numbers or identifiers

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

SequentialGroup

-0.43

imas

-0.40

hmig

-0.38

Ƹ

-0.38

elsen

-0.38

 relieved

-0.37

RenderAtEndOf

-0.37

typeDefinition

-0.36

 اش

-0.36

 returned

-0.36

POSITIVE LOGITS

 HasFactory

0.66

 AssemblyVersion

0.63

LookAnd

0.59

ViewFeatures

0.59

FontOfSize

0.51

apiVersion

0.50

#+#

0.50

AutoScaleMode

0.49

styleable

0.48

prefixer

0.48

Activations Density 0.003%

elements related to API versioning and specifications

version numbers or identifiers

No Comments

No Known Activations