INDEX

Explanations

phrases related to data analysis and evaluation

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

of their, the search, the event, the global

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

RectangleBorder

-0.53

 poveznice

-0.42

ніципа

-0.40

 CanadaChoose

-0.40

 Taktlose

-0.38

 Мексичка

-0.36

 betweenstory

-0.36

henswürdigkeiten

-0.36

цездатний

-0.35

SwingConstants

-0.35

POSITIVE LOGITS

 AssemblyCulture

0.58

 Drain

0.50

InputTagHelper

0.50

 Situs

0.49

pannt

0.47

 Sender

0.47

férences

0.47

DateField

0.46

𓇠

0.46

 Perubahan

0.46

Activations Density 1.146%

phrases related to data analysis and evaluation

of their, the search, the event, the global

No Comments

No Known Activations