INDEX

Explanations

phrases related to time references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

Rolf

-0.80

']?>

-0.72

 Balboa

-0.70

 Rolf

-0.69

Bibliographie

-0.69

 EconPapers

-0.68

 Paxton

-0.67

 Beatrix

-0.66

 iParam

-0.61

 Inca

-0.61

POSITIVE LOGITS

during

1.63

 During

1.57

During

1.51

 during

1.48

 DURING

1.45

durante

1.44

 Durante

1.28

 durant

1.23

Durante

1.21

 tijdens

1.20

Activations Density 0.107%

phrases related to time references

No Comments

No Known Activations