INDEX

Explanations

occurrences of the word "sand" and its variations

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

sand or sand-related names

np_acts-logits-general · gemini-2.5-flash-lite

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 Woche

-0.46

ofType

-0.46

 følge

-0.45

 wireType

-0.43

ppio

-0.42

 Motive

-0.41

 ocorrer

-0.40

Willy

-0.39

 пона

-0.39

期

-0.39

POSITIVE LOGITS

 Sand

1.24

Sand

1.14

 sand

1.11

 SAND

1.05

 Sandberg

0.97

 Sandler

0.92

 Sands

0.91

SAND

0.90

 Sanderson

0.89

 Shand

0.88

Activations Density 0.007%

occurrences of the word "sand" and its variations

sand or sand-related names

No Comments

No Known Activations