INDEX

Explanations

numerical values or statistics related to significant concepts.

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

numerical values interspersed with specific terms and names.

oai_token-act-pair · gpt-3.5-turbo Triggered by @bot

prepositions, particularly "von", "aus", "in", "mit", and "bei" in German text.

oai_token-act-pair · claude-3-5-sonnet-20240620 Triggered by @bot

Following the words "von," "dem," "der," or "den"

np_acts-logits-general · gemini-2.0-flash

"von" followed by names

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_20/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

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

 Monfieur

-0.86

Efq

-0.82

 itſelf

-0.82

 himſelf

-0.80

 Majefty

-0.79

 themſelves

-0.78

 myſelf

-0.78

 alſo

-0.77

 faſt

-0.76

 pleaſure

-0.75

POSITIVE LOGITS

 einer

0.78

dem

0.77

 einem

0.76

der

0.75

den

0.55

()){

0.52

 divers

0.49

MLLoader

0.48

 mehreren