INDEX

Explanations

declarations or statements, particularly those that conclude a thought or text

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

multilingual programming or data errors

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

Negative Logits

lop

-0.47

LG

-0.44

AdapterView

-0.42

 resto

-0.42

 resc

-0.41

def

-0.41

bog

-0.40

Niv

-0.40

vcf

-0.40

nod

-0.40

POSITIVE LOGITS

MemoryWarning

0.68

 auroit

0.60

 يتيمه

0.56

 avoient

0.55

 StatefulWidget

0.54

 enfans

0.54

 zijne

0.53

 étoient

0.53

 feroit

0.53

StoryboardSegue

0.50

Activations Density 0.001%

declarations or statements, particularly those that conclude a thought or text

multilingual programming or data errors

No Known Activations