INDEX

Explanations

terms related to healthcare and professional standards

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

numbers and units or symbols

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-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

-0.57

ſelves

-0.54

 myſelf

-0.52

 fubject

-0.52

 Conſ

-0.49

 للمعارف

-0.49

 AssemblyVersion

-0.48

 Eilish

-0.47

ſelf

-0.47

 whoſe

-0.47

POSITIVE LOGITS

 nakalista

0.44

 <>",

0.43

complexType

0.38

SpringRunner

0.36

 ویکی‌پدیای

0.36

apter

0.36

 HasFactory

0.36

SharedModule

0.35

 sanitaires

0.34

 jaket

0.34

Activations Density 0.049%

terms related to healthcare and professional standards

numbers and units or symbols

No Comments

No Known Activations