INDEX

Explanations

attributes related to product features and specifications

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

X followed by Y

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

 betweenstory

-0.68

DockStyle

-0.66

 ब्रेकडाउन

-0.58

 Meksiku

-0.57

 ModelExpression

-0.56

 propOrder

-0.54

новништво

-0.50

berdayakan

-0.50

LookAnd

-0.50

 linkovi

-0.49

POSITIVE LOGITS

 itself

0.40

 yourself

0.39

 himself

0.37

󠁴

0.36

 ourselves

0.35

 instead

0.35

itself

0.35

 szere

0.35

έχ

0.34

ModelAdmin

0.34

Activations Density 0.022%

attributes related to product features and specifications

X followed by Y

No Comments

No Known Activations