INDEX

Explanations

expressions related to evaluation and analysis of various groups or entities

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

"... as a whole."

np_acts-logits-general · gemini-2.0-flash

as a whole

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_22/width_16k/average_l0_72

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 виправивши

-0.58

angeran

-0.51

 referenties

-0.50

 posibil

-0.49

vací

-0.49

 mukana

-0.48

alitions

-0.48

TintMode

-0.48

čník

-0.48

offensive

-0.48

POSITIVE LOGITS

 whole

1.91

whole

1.75

Whole

1.56

 Whole

1.49

 WHOLE

1.44

 geheel

0.99

 целом

0.97

 entirety

0.97

 entire

0.96

 ganzen

0.91

Activations Density 0.198%

expressions related to evaluation and analysis of various groups or entities

"... as a whole."

as a whole

No Comments

No Known Activations