INDEX

Explanations

key terms and phrases related to statistical analysis and medical conditions

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

only if contrasted

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

Negative Logits

httphttps

-0.68

 noqa

-0.53

nologue

-0.53

rsiniz

-0.52

rogly

-0.52



-0.50

político

-0.50

 CURIAM

-0.49

Alike

-0.49

AddTagHelper

-0.49

POSITIVE LOGITS

 only

0.41

0.32

ThroughAttribute

0.32

frem

0.30

 selective

0.29

0.27

 seulement

0.27

tran

0.26

 recreate

0.26

Activations Density 0.119%

key terms and phrases related to statistical analysis and medical conditions

only if contrasted

No Known Activations