INDEX

Explanations

references to specific technical terms or concepts in a formal context

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

citations and references

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

 ModelExpression

-0.29

=?,

-0.28

 fans

-0.27

 Benutz

-0.27

スポンサーリンク

-0.26

 aggressive

-0.26

 alue

-0.26

 agresión

-0.26

 gastronomía

-0.25

ativement

-0.25

POSITIVE LOGITS

uxxxx

0.80

webElementXpaths

0.72

цездатний

0.70

StoryboardSegue

0.65

 informée

0.60

tanleria

0.60

 autorytatywna

0.60

лтамалар

0.60

MockMvc

0.60

IUrlHelper

0.59

Activations Density 0.004%

references to specific technical terms or concepts in a formal context

citations and references

No Comments

No Known Activations