INDEX

Explanations

terms related to pathogen dynamics and interactions with hosts

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

infectivity, recovery, mortality rates

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

pecabe

-0.60

accompag

-0.59

 snippetHide

-0.59

ðsíða

-0.57

majánló

-0.56

 NavController

-0.54

 femininos

-0.52

 Händen

-0.52

zbęd

-0.51

 témoig

-0.50

POSITIVE LOGITS

LabelTagHelper

0.52

 cura

0.35

yyr

0.33

acc

0.32

🎳

0.31

Rhestr

0.30

>{{

0.29

SANTA

0.29

={()

0.29

cap

0.29

Activations Density 0.266%

terms related to pathogen dynamics and interactions with hosts

infectivity, recovery, mortality rates

No Comments

No Known Activations