INDEX

Explanations

references to processes involving analysis and serving in scientific or procedural contexts

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

ValueStyle

-0.66

 محفوظة

-0.51

 يتيمه

-0.49

})`

-0.49

 Propaganda

-0.48

 вікісторінку

-0.47

<!--

-0.47

 Suspension

-0.46

 localObject

-0.45

ACI

-0.45

POSITIVE LOGITS

GEBURTSDATUM

0.59

invokeLater

0.44

routeProvider

0.43

 AssemblyCulture

0.43

دانشنامهٔ

0.42

 henvisninger

0.42

MessageTagHelper

0.40

RotationOrder

0.40

 aapt

0.38

ClientSize

0.36

Activations Density 0.190%

references to processes involving analysis and serving in scientific or procedural contexts

No Comments

No Known Activations