INDEX

Explanations

phrases related to therapeutic applications and efficacy in biomedical research

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

in vivo

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

 препратки

-0.64

Datuak

-0.60

 nahilalakip

-0.57

 ſta

-0.55

ſelves

-0.54

 ſtand

-0.54

 Jefus

-0.53

 itſelf

-0.52

يميديا

-0.50

 myſelf

-0.50

POSITIVE LOGITS

WithMany

0.40

RRect

0.39

Kur

0.39

 numerus

0.39

 real

0.38

Whi

0.37

 Wish

0.37

Rptr

0.37

 Charter

0.36

 Groom

0.36

Activations Density 0.421%

phrases related to therapeutic applications and efficacy in biomedical research

in vivo

No Comments

No Known Activations