INDEX

Explanations

technical specifications and features related to products

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

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

ebenarnya

-0.28

 Serikat

-0.28

 Accesat

-0.28

 Krakowie

-0.28

englanniksi

-0.27

éně

-0.25

 public

-0.25

 Gefühle

-0.25

속

-0.24

 naudoti

-0.24

POSITIVE LOGITS

WriteTagHelper

0.81

 disambiguazione

0.71

fjspx

0.70

tagHelperRunner

0.61

Personendaten

0.59

0.58

 beginnetje

0.58

migrationBuilder

0.58

 AppDelegate

0.57

WithIOException

0.57

Activations Density 0.025%

technical specifications and features related to products

No Comments

No Known Activations