INDEX

Explanations

references to organizations and technical terminologies

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

dejal.com

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

des

-0.33

ctrl

-0.32

TacToe

-0.29

Addresses

-0.29

Asset

-0.29

skat

-0.29

 upsetting

-0.29

 adopté

-0.28

 upsets

-0.28

 neque

-0.28

POSITIVE LOGITS

 виправивши

0.73

出版年

0.71

RTLI

0.66

ThroughAttribute

0.65

hoeddwyd

0.62

Jeografia

0.62

 >=",

0.59

ftagPool

0.59

AddHtmlAttribute

0.57

PYX

0.56

Activations Density 0.178%

references to organizations and technical terminologies

dejal.com

No Comments

No Known Activations