INDEX

Explanations

statistical data related to percentages or numerical comparisons.

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

details related to scientific data and measurements.

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

The main thing this neuron does is find adverbs and phrases indicating comparison or relative positioning, particularly "respectively".

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

"respectively"

np_acts-logits-general · gemini-2.0-flash

respectively,

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

Personendaten

-0.90

 Roskov

-0.82

*/,

-0.78

tvguidetime

-0.78

}',

-0.78

',

-0.77

 nahilalakip

-0.76

 beginnetje

-0.76

*/,

-0.75

 disambiguazione

-0.75

POSITIVE LOGITS

1.05

TagHelpers

0.57

StoreMessageInfo

0.43

Trama

0.43

0.41

 prochains

0.39

 että

0.38

 Thron

0.38

ảy

0.38

klo

0.37

Activations Density 0.457%

statistical data related to percentages or numerical comparisons.

details related to scientific data and measurements.

The main thing this neuron does is find adverbs and phrases indicating comparison or relative positioning, particularly "respectively".

"respectively"

respectively,

No Comments

No Known Activations