INDEX

Explanations

references to diagnostic accuracy, specifically related to false positives and false negatives in testing

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

positive or negative outcomes

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 9-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_9/width_131k/average_l0_121

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.9.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

 nhàng

-0.29

<?>

-0.26

log

-0.26

 anzunehmen

-0.25

ด

-0.25

卖家

-0.24

onomian

-0.24

TaskList

-0.24

 sentimentos

-0.24

 rest

-0.24

POSITIVE LOGITS

findpost

0.84

########.

0.76

TagMode

0.73

windowFixed

0.71

 defaultstate

0.68

 nahilalakip

0.68

+:+

0.68

rungsseite

0.68

 مرئيه

0.65

 nakalista

0.65

Activations Density 0.410%

references to diagnostic accuracy, specifically related to false positives and false negatives in testing

positive or negative outcomes

No Comments

No Known Activations