INDEX

Explanations

numerical values and expressions related to sections or parts of documents or locations

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

+:+

-0.44

excellents

-0.39

 tools

-0.38

tangentMode

-0.36

 simplicité

-0.35

 Simplicity

-0.35

🇶

-0.34

Simplicity

-0.34

 simplicity

-0.32

 flagship

-0.32

POSITIVE LOGITS

 الحره

0.46

 propOrder

0.46

 BoxFit

0.45

KURZBESCHREIBUNG

0.44

 nunc

0.43

findpost

0.43

niająca

0.42

 كومونز

0.41

 rayures

0.41

 jadx

0.40

Activations Density 0.055%

numerical values and expressions related to sections or parts of documents or locations

No Comments

No Known Activations