INDEX

Explanations

numerical values and their context related to medical or behavioral outcomes

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

influence and affect

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

 AssemblyTitle

-0.46

TestingModule

-0.43

expandindo

-0.41

reactstrap

-0.40

enschutz

-0.40

 bezw

-0.40

anyaan

-0.40

 Schlu

-0.39

 AssemblyCompany

-0.39

 tempfile

-0.38

POSITIVE LOGITS

 influencing

0.49

 afect

0.42



0.40

 influences

0.39

 beeinf

0.39

 verändert

0.38

 influence

0.38

 significantly

0.38

 afecta

0.37

 Affects

0.37

Activations Density 0.126%

numerical values and their context related to medical or behavioral outcomes

influence and affect

No Comments

No Known Activations