INDEX

Explanations

references to information sources and availability

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

details and information

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

InjectAttribute

-0.74

addCriterion

-0.67

ValueStyle

-0.66

новништво

-0.62

addContainerGap

-0.61

SharedDtor

-0.60

matchCondition

-0.59

GEBURTSDATUM

-0.59

GEBURTS

-0.58

LLocation

-0.58

POSITIVE LOGITS

 detalles

0.40

nestjs

0.39

 announced

0.39

 details

0.37

くわ

0.35

 iliş

0.35

 detailed

0.34

 information

0.34

 info

0.33

 projeto

0.33

Activations Density 0.008%

references to information sources and availability

details and information

No Comments

No Known Activations