INDEX

Explanations

words that indicate ongoing or repeated actions or events in a narrative context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 10-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_10/width_16k/average_l0_77

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.10.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

-0.75

 Rolf

-0.68

 Beatrix

-0.67

 propOrder

-0.65

 Balboa

-0.65

']?>

-0.63

 iParam

-0.59

 Inca

-0.58

 creatinine

-0.57

pulumi

-0.57

POSITIVE LOGITS

during

1.47

 During

1.43

During

1.35

durante

1.34

 DURING

1.32

 during

1.30

 Durante

1.14

 durant

1.11

Durante

1.08

 periods

1.05

Activations Density 0.083%

words that indicate ongoing or repeated actions or events in a narrative context

No Comments

No Known Activations