INDEX

Explanations

programming-related keywords and syntax

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_33/width_131k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.33.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

 propOrder

-0.47

 express

-0.37

 rocket

-0.34

rocket

-0.34

Express

-0.34

Rocket

-0.33

 expressed

-0.33

棘

-0.31

Wa

-0.31

 expri

-0.31

POSITIVE LOGITS

MUD

0.66

MUD

0.65

mud

0.64

mud

0.61

TagHelper

0.58

Mud

0.57

Mud

0.57

NPC

0.56

NPC

0.54

 NPCs

0.54

Activations Density 0.028%

programming-related keywords and syntax

No Comments

No Known Activations