INDEX

Explanations

expressions related to performance evaluation and qualification

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

qualifying performance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 16-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_16/width_65k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.16.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

ⓧ

-0.57

addGap

-0.55

Pyx

-0.55

Filmografie

-0.55

ACHE

-0.51

">//

-0.51

 jährlich

-0.50

ждународ

-0.50

 distritos

-0.50

 vastaan

-0.49

POSITIVE LOGITS

TagMode

0.65

 himſelf

0.63

 Thursday

0.61

RegistryLite

0.61

 Friday

0.60

 Wednesday

0.60

stood

0.59

 myſelf

0.58

 morning

0.58

Thursday

0.57

Activations Density 0.015%

expressions related to performance evaluation and qualification

qualifying performance

No Comments

No Known Activations