INDEX

Explanations

programming-related terms and structures in code snippets

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

fjspx

-0.54

 GoogleFonts

-0.48

 blij

-0.46

principalTable

-0.45

ApiProperty

-0.43

Mv

-0.43

دانشنامهٔ

-0.42

folgen

-0.41

bentar

-0.41

期刊论文

-0.41

POSITIVE LOGITS

 gynhyrchwyd

0.43

 Infórmanos

0.40

ArgumentParser

0.40

elemField

0.40

 PeEnEo

0.37

 useAppContext

0.37

 Anfitrión

0.37

 calculées

0.37

spraak

0.35

interpr

0.35

Activations Density 0.043%

programming-related terms and structures in code snippets

No Comments

No Known Activations