INDEX

Explanations

punctuation marks and sentence-ending elements, particularly commas and periods.

oai_token-act-pair · claude-3-5-sonnet-20240620 Triggered by @myinzi123

elements of conversational structure and punctuation usage

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

Non-English punctuation

np_acts-logits-general · gemini-2.0-flash

multilingual punctuation and technical terms

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-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 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

 egregious

-0.54

 supplemental

-0.53

 federally

-0.52

OVERR

-0.52

tibus

-0.50

 someplace

-0.49

 resiliency

-0.48

 congressional

-0.48

 cappuccio

-0.46

 Jackson

-0.45

POSITIVE LOGITS

 disambiguazione

0.63

 للاسماء

0.62

adpleegd

0.61

بوابة

0.61

tagHelperRunner

0.60

 tartalomajánló

0.57

 незавершена

0.55

 DBNull

0.51

usercontent

0.50

fjspx