INDEX

Explanations

indicators of code changes or modifications in programming files

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

proto Message snippet XML regular expression

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

кота

-0.35

-0.34

-0.32

 judgement

-0.28

 Ebene

-0.27

 judgements

-0.27

 wrap

-0.26

 Zink

-0.26

POSITIVE LOGITS

httphttps

1.01

 autorytatywna

1.00

 defaultstate

0.93

 Roskov

0.80

LookAnd

0.80

تقاوى

0.78

 nahilalakip

0.77

 незавершена

0.77

 kasarigan

0.76

الحياه

0.76

Activations Density 0.001%

indicators of code changes or modifications in programming files

proto Message snippet XML regular expression

No Comments

No Known Activations