INDEX

Explanations

references to biological or medical processes and their metrics

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

malfunctions or errors

np_acts-logits-general · gemini-2.0-flash

medical and technical terms

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

ülerin

-0.42

Virtual

-0.38

the

-0.38

 transición

-0.36

 four

-0.36

 highly

-0.36

Pregnancy

-0.35

 very

-0.35

żo

-0.35

 abzu

-0.34

POSITIVE LOGITS

ReusableCell

0.66

+#+

0.66

SizeMode

0.66

 ostavi

0.64

InitVars

0.63

Билгалдахарш

0.63

 Exacts

0.63

 saites

0.62

 consultato

0.61

Erstellt

0.60

Activations Density 1.745%

references to biological or medical processes and their metrics

malfunctions or errors

medical and technical terms

No Comments

No Known Activations