INDEX

Explanations

terms related to programming tasks or operational metrics

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

user queries with code

np_acts-logits-general · gemini-2.5-flash-lite

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 fallu

-0.34

asc

-0.29

 verrez

-0.27

Photo

-0.27

onAdd

-0.26

 info

-0.25

Upper

-0.25

wiſe

-0.25

 impri

-0.25

Seen

-0.24

POSITIVE LOGITS

coroutines

2.86

awaiter

0.99

WaitGroup

0.95

coroutine

0.93

Coroutine

0.93

 asyncio

0.85

IntoConstraints

0.76

ruptedException

0.75

complexContent

0.75

 للاسماء

0.75

Activations Density 0.000%

terms related to programming tasks or operational metrics

user queries with code

No Comments

No Known Activations