INDEX

Explanations

terms related to mathematical concepts and structures, particularly those involving embeddings and measures

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

Followed by a two or three letter token

np_acts-logits-general · gemini-2.0-flash

Carleson, Frechet, birational, stokeslet beginnings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

</strong>

-0.51

’

-0.49

</em>

-0.47

…

-0.46

***

-0.45

‘

-0.44

‘

-0.44

 opening

-0.41

POSITIVE LOGITS

 utafitiHapana

1.04

Personendaten

0.96

ArrowToggle

0.90

 CURIAM

0.90

 AssemblyCulture

0.87

ſelf

0.86

RenderAtEndOf

0.85

InputTagHelper

0.84

neſs

0.83

 wireType

0.83

Activations Density 1.535%

terms related to mathematical concepts and structures, particularly those involving embeddings and measures

Followed by a two or three letter token

Carleson, Frechet, birational, stokeslet beginnings

No Comments

No Known Activations