INDEX

Explanations

specific numerical or quantified references

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

technical and scientific terms

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

Referencie

-0.36

ttet

-0.34

 renou

-0.34

 informée

-0.32

 مُعرِّف

-0.32

 Abonnez

-0.32

 interpretations

-0.32

 Moscú

-0.30

 koja

-0.30

Välislingid

-0.30

POSITIVE LOGITS

 '\\;'

0.68

 يتيمه

0.67

UrlResolution

0.65

 Wikispecies

0.64

fortawesome

0.63

SequentialGroup

0.62

\{\\

0.61

وفاته

0.61

UnitTesting

0.59

ChildScrollView

0.56

Activations Density 0.083%

specific numerical or quantified references

technical and scientific terms

No Comments

No Known Activations