INDEX

Explanations

quantitative relationships or data that indicate phenomena occurring over time.

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

information related to legal proceedings and investigations.

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

numbered or lettered list items within parentheses, often used to structure legal or technical documents.

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

Opening parentheses followed by a number

np_acts-logits-general · gemini-2.0-flash

enumerations like (a), (i), (1)

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Wikimedijinoj

-1.00

 Paglinawan

-0.94

 AssemblyProduct

-0.88

parsedMessage

-0.85

contentLoaded

-0.83

 незавершена

-0.81

abestanden

-0.81

 Chwiliwch

-0.80

WriteTagHelper

-0.80

 betweenstory

-0.79

POSITIVE LOGITS

1.21

1.14

1.02

0.98

0.86

0.81

0.77

а

0.75

①

0.74

⑴

0.73

Activations Density 0.665%