INDEX

Explanations

phrases indicating recent actions or events

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 22-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_22/width_16k/average_l0_123

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.22.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

 nantinya

-0.40

 entière

-0.40

 násled

-0.37

 różne

-0.36

 eventually

-0.36

 ulike

-0.35

 awalnya

-0.35

 quelcon

-0.35

 później

-0.35

 począ

-0.35

POSITIVE LOGITS

 recently

0.91

recently

0.89

Recently

0.85

 Recently

0.82

 недавно

0.80

 recentemente

0.78

 freshly

0.76

 recientemente

0.75

 Newly

0.73

newly

0.73

Activations Density 0.215%

phrases indicating recent actions or events

No Comments

No Known Activations