INDEX

Explanations

references to researchers and authors indicated by a "J" followed by a number

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

Followed by a letter J

np_acts-logits-general · gemini-2.0-flash

name initials

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

StreetMap

-0.82

erdings

-0.80

httphttps

-0.72

TagMode

-0.72

amblea

-0.65

ympto

-0.64

pergillus

-0.64

elemField

-0.64

:✨

-0.63

اء

-0.63

POSITIVE LOGITS

0.63

tached

0.55

xticks

0.47

Aftermath

0.44

 kubwa

0.44

Bronnen

0.43

LOADED

0.43

áctica

0.43

ModelForm

0.41

'])->

0.41

Activations Density 0.191%

references to researchers and authors indicated by a "J" followed by a number

Followed by a letter J

name initials

No Comments

No Known Activations