INDEX

Explanations

proper nouns, especially names of authors and their affiliations

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

before citations and names

np_acts-logits-general · gemini-2.0-flash

Persian/Arabic names and titles

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 24-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_24/width_16k/average_l0_73

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.24.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

TestingModule

-0.53

 face

-0.47

thew

-0.46

축

-0.45

quently

-0.44

getWriter

-0.44

 Illuminate

-0.44

aufbau

-0.42

 černá

-0.42

θ

-0.42

POSITIVE LOGITS

PyExc

0.68

MemoryWarning

0.63

 bezeichneter

0.61

poor

0.60

 POUR

0.60

 ProtoMessage

0.60

 Poor

0.59

Poor

0.59

таратура

0.58

ContentAsync

0.58

Activations Density 0.153%

proper nouns, especially names of authors and their affiliations

before citations and names

Persian/Arabic names and titles

No Comments

No Known Activations