INDEX

Explanations

elements and attributes related to biological classifications and processes

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

gene level, type, category

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

URLException

-0.54

-0.39

range

-0.39

kym

-0.38

 handled

-0.37

blico

-0.37

Geplaatst

-0.37

EndContext

-0.37

 discussing

-0.36

討論

-0.36

POSITIVE LOGITS

 autorytatywna

0.45

 Chwiliwch

0.44

SharedCtor

0.44

 kasarigan

0.42

Personendaten

0.40

LOYEES

0.40

araoh

0.39

 FetchType

0.38

WireFormat

0.38

 initComponents

0.38

Activations Density 0.177%

elements and attributes related to biological classifications and processes

gene level, type, category

No Comments

No Known Activations