INDEX

Explanations

numerical data, especially related to software versioning and timestamps

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

%s placeholders and durations

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 AssemblyTitle

-0.40

람

-0.35

Keys

-0.34

dė

-0.34

 Foer

-0.34

Demografie

-0.33

Fordítás

-0.33

<blockquote>

-0.33

joit

-0.33

pośred

-0.32

POSITIVE LOGITS

 disambiguazione

0.63

 GenerationType

0.62

 متعلقه

0.60

XmlAccessorType

0.60

----</

0.60

addPreferredGap

0.60

 ujednoznacz

0.57

FetchType

0.57

 JpaRepository

0.57

حياته

0.57

Activations Density 0.032%

numerical data, especially related to software versioning and timestamps

%s placeholders and durations

No Comments

No Known Activations