INDEX

Explanations

references to documentation or guides related to software and programming

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

documentation and manuals

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

 BoxFit

-0.53

bootstrapcdn

-0.49

EndContext

-0.47

HEC

-0.44

 tracksuit

-0.42

 nanny

-0.37

TNT

-0.37

PSE

-0.37

nationality

-0.36

TNT

-0.36

POSITIVE LOGITS

rungsseite

0.49

 explicado

0.47

 disambiguazione

0.45

 ujednoznacz

0.45

DebuggerStep

0.45

 dorada

0.45

uș

0.44

Carriera

0.44

elezo

0.43

 refroid

0.43

Activations Density 0.087%

references to documentation or guides related to software and programming

documentation and manuals

No Comments

No Known Activations