INDEX

Explanations

references to theft or loss

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

lost, stolen, or missing items

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

 snippetHide

-0.46

thage

-0.45

WebMethod

-0.42

DOCTYPE

-0.41

 labored

-0.40

diz

-0.40

holl

-0.39

цездатний

-0.38

 observable

-0.38

writerow

-0.37

POSITIVE LOGITS

 LOST

0.64

 Lost

0.63

 lost

0.60

Lost

0.60

lost

0.57

Losing

0.57

 pérdida

0.57

 Losing

0.56

LOST

0.54

 perdida

0.52

Activations Density 0.011%

references to theft or loss

lost, stolen, or missing items

No Comments

No Known Activations