INDEX

Explanations

words related to specific food preparation and artisan activities

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

activities and locations

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

ROIT

-0.36

MessageTagHelper

-0.36

zsche

-0.36

chieht

-0.35

 paio

-0.35

urlencoded

-0.35

 hipó

-0.34

 possesso

-0.33

 Burch

-0.33

nalités

-0.33

POSITIVE LOGITS

 âgées

0.54

Geplaatst

0.53

 مشين

0.52

endpush

0.49

 تضيفلها

0.47

TagMode

0.47

DBNull

0.46

Belt

0.45

MessageOf

0.44

WriteBarrier

0.43

Activations Density 0.013%

words related to specific food preparation and artisan activities

activities and locations

No Comments

No Known Activations