INDEX

Explanations

comments and documentation in code

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

documentation comments

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

 čierna

-0.57

 Einfluß

-0.52

 bílá

-0.50

ɵɵelement

-0.49

 برانيه

-0.48

 sufrimiento

-0.47

protetor

-0.46

miyorum

-0.46

 wszyst

-0.45

 plástica

-0.45

POSITIVE LOGITS

VersionUID

0.50

 HasFactory

0.47

 ujednoznacz

0.45

 title

0.44

imageshack

0.42

 AssemblyCompany

0.41

 Souha

0.38

jooq

0.38

Kaieteur

0.37

Fly

0.36

Activations Density 0.005%

comments and documentation in code

documentation comments

No Comments

No Known Activations