INDEX

Explanations

subtle indications of medical conditions and their diagnosis or effects

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

Subtle, hidden, or undetected phenomena

np_acts-logits-general · gemini-2.0-flash

subclinical hidden conditions

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

PRIORITY

-0.33

narrow

-0.30

DISP

-0.28

 babak

-0.28

 Plummer

-0.27

 resort

-0.27

 exclusivamente

-0.27

PREVIEW

-0.26

 sport

-0.26

 NgModule

-0.26

POSITIVE LOGITS

 '\\;'

1.04

 hidden

0.88

 secretly

0.82

 silently

0.82



0.82

 Hidden

0.76

 محفوظة

0.76

 quietly

0.76

hidden

0.71

 invisible

0.71

Activations Density 0.479%

subtle indications of medical conditions and their diagnosis or effects

Subtle, hidden, or undetected phenomena

subclinical hidden conditions

No Comments

No Known Activations