INDEX

Explanations

terms related to confidence measures and statistical analysis

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

confidence intervals, protective coating, selection transistor

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

 EconPapers

-0.49

 propOrder

-0.48

 başına

-0.47

 quæ

-0.46

 avoient

-0.44

icznego

-0.44

 getNome

-0.43

affaire

-0.43

icznie

-0.43

numerusform

-0.42

POSITIVE LOGITS

 Hivatalos

0.44

},{

0.44

}]

0.42

 pole

0.41

úgó

0.40

日閲覧

0.38

 action

0.38

 leeren

0.38

 Cellulose

0.38

 stir

0.38

Activations Density 0.155%

terms related to confidence measures and statistical analysis

confidence intervals, protective coating, selection transistor

No Comments

No Known Activations