INDEX

Explanations

elements related to user interface components in a coding context

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

Code snippets using "footer" or "Highlight"

np_acts-logits-general · gemini-2.0-flash

closing tags

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

multer

-0.69

 AssemblyCompany

-0.69

ConstraintMaker

-0.66

TestingModule

-0.65

<unused74>

-0.64

<unused41>

-0.64

<unused8>

-0.63

<unused16>

-0.63

[@BOS@]

-0.63

<pad>

-0.63

POSITIVE LOGITS

 everything

0.41

enterOuterAlt

0.36

 Everything

0.34

Everything

0.31

Bronnen

0.31

everything

0.30

STRUCTIONS

0.30

SizedBox

0.29

hawatir

0.28

 supporting

0.28

Activations Density 0.028%

elements related to user interface components in a coding context

Code snippets using "footer" or "Highlight"

closing tags

No Comments

No Known Activations