INDEX

Explanations

technical terms and jargon related to molecular electronics and their applications

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

before numbers or symbols

np_acts-logits-general · gemini-2.0-flash

descriptive modifiers

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

笈

-0.39

함

-0.38

 exception

-0.37

 except

-0.35

Brin

-0.35

crollView

-0.34

 nakalista

-0.34

addPreferredGap

-0.34

septic

-0.34

 Except

-0.34

POSITIVE LOGITS



0.49

 Newswire

0.47

Autoritní

0.46

 AssemblyCulture

0.44

verifyException

0.41

 ivelany

0.40

 Drapeau

0.40

tagext

0.40

Dtor

0.39

رشف

0.39

Activations Density 2.506%

technical terms and jargon related to molecular electronics and their applications

before numbers or symbols

descriptive modifiers

No Comments

No Known Activations