INDEX

Explanations

references to the word "saga."

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

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

(\

-0.42

 Utilizamos

-0.41

Tooth

-0.38

Teeth

-0.38

TRAILING

-0.36

paddingBottom

-0.36

Fij

-0.36

((

-0.36

POSITIVE LOGITS

 Saga

1.21

Saga

1.13

 saga

1.12

saga

1.09

 Odyssey

0.72

yssey

0.70

 adventure

0.70

 odyssey

0.69

 adventures

0.69

 Chapters

0.69

Activations Density 0.002%

references to the word "saga."

No Comments

No Known Activations