INDEX

Explanations

elements related to treatments and clinical procedures

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

construct

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

RTSN

-0.33

À

-0.32

Answer

-0.31

Title

-0.30

 MessageBoxIcon

-0.29

-0.28

 atve

-0.28

上

-0.28

<?

-0.28

openSession

-0.28

POSITIVE LOGITS

StoryboardSegue

0.81

OGND

0.75

Personendaten

0.74

 Paglinawan

0.70

◹

0.67

RenderAtEndOf

0.65

évaluateur

0.61

RTEE

0.60

 intStringLen

0.59

Viitteet

0.59

Activations Density 0.076%

elements related to treatments and clinical procedures

construct

No Comments

No Known Activations