INDEX

Explanations

comparative measurements related to size, performance, and efficiency in various contexts

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

InjectAttribute

-0.52

 resourceCulture

-0.52

EClass

-0.47

AsUp

-0.46

BagConstraints

-0.46

 Савезне

-0.44

 pinulongan

-0.43

 onlyOwner

-0.43

 setuptools

-0.43

AccessorTable

-0.42

POSITIVE LOGITS

BeginContext

0.53

 AssemblyVersion

0.51

 averaging

0.41

LabelTagHelper

0.40

 nämlich

0.38

 onOptions

0.38

fjspx

0.37

 heiligen

0.36

soort

0.35

 hiszen

0.35

Activations Density 0.129%

comparative measurements related to size, performance, and efficiency in various contexts

No Comments

No Known Activations