INDEX

Explanations

terms related to overhead costs and false positives in processes

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

learning curve, precedents, false positive, overhead

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

Pyx

-0.43

astify

-0.39

canestro

-0.38

vistar

-0.37

Portail

-0.37

OfYear

-0.36

djangoproject

-0.36

 Portail

-0.35

 Animal

-0.35

getActivity

-0.35

POSITIVE LOGITS

 precedents

0.52

RenderAtEndOf

0.52

 chistes

0.52

leneck

0.50

Prerequisite

0.48

 hectáreas

0.48

 sorpresas

0.48

nextProps

0.47



0.47

 kink

0.46

Activations Density 0.063%

terms related to overhead costs and false positives in processes

learning curve, precedents, false positive, overhead

No Comments

No Known Activations