INDEX

Explanations

references to versioning in software or documentation

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

revision and assembly information

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

 schoenen

-0.52

 scattata

-0.52

MathML

-0.50

 dissatisfied

-0.48

 unhappy

-0.46

 meisjes

-0.46

échange

-0.45

IBOutlet

-0.45

ThroughAttribute

-0.45

 immorality

-0.44

POSITIVE LOGITS

 AssemblyCulture

1.12

 revision

0.73

 Revision

0.66

Revision

0.64

revision

0.62

 revisions

0.52

Demografie

0.50

protoimpl

0.49

inaison

0.49

 kasarigan

0.48

Activations Density 0.000%

references to versioning in software or documentation

revision and assembly information

No Comments

No Known Activations