INDEX

Explanations

structured data formats, specifically looking for various delimiters and data types

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 4-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_4/width_16k/average_l0_124

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.4.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

 Parce

-0.61

 Monfieur

-0.60

 ――――

-0.58

Efq

-0.57

 Stodd

-0.55

})));

-0.55

 condensate

-0.53

 Fusarium

-0.52

 excuses

-0.51

])));

-0.51

POSITIVE LOGITS

 Walkover

0.78

Autoritní

0.73

Enllaces

0.67

сылкі

0.65

Personensuche

0.63

 arşivlendi

0.62

וויק

0.60

 disambiguazione

0.58

marvin

0.58

pieler

0.57

Activations Density 0.130%

structured data formats, specifically looking for various delimiters and data types

No Comments

No Known Activations