INDEX

Explanations

key terms indicating intent or purpose

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

with an aim to

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

blico

-0.57

󠁴

-0.56

outWeight

-0.53

Gambas

-0.53

 estekak

-0.51

 IFTT

-0.50

thâu

-0.50

мум

-0.50

 BorderSide

-0.50

хівовано

-0.49

POSITIVE LOGITS

ंदीखरीदारी

0.42

सन्दर्भ

0.39

Witam

0.35

tagHelperRunner

0.34

 AssemblyCompany

0.34

TypedDataSet

0.33

Odkazy

0.31

 informée

0.30

 šal

0.30

énieur

0.29

Activations Density 0.000%

key terms indicating intent or purpose

with an aim to

No Comments

No Known Activations