INDEX

Explanations

the occurrence of specific code implementation patterns or structures

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

UrlResolution

-0.67

 typelib

-0.58

 Autorizaciones

-0.56

transQ

-0.56

 يتيمه

-0.55

inSlope

-0.55

 صوتيه

-0.53

 ویکی‌پدی

-0.53

astify

-0.51

elemField

-0.50

POSITIVE LOGITS

CodeDom

0.49

Erstellt

0.47

ViewFeatures

0.45

 kasarigan

0.42

AccessorTable

0.42

PerformLayout

0.41

httphttps

0.41

 recueil

0.41

 OnInit

0.41

Ar

0.39

Activations Density 0.002%

the occurrence of specific code implementation patterns or structures

No Comments

No Known Activations