INDEX

Explanations

references to translation in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

 ventilación

-0.46

 plegable

-0.43

 Lingkungan

-0.43

prü

-0.41

 compound

-0.41

 Constitucional

-0.40

yttö

-0.40

 payroll

-0.39

 weather

-0.39

 rounds

-0.39

POSITIVE LOGITS

rim

0.71

Rim

0.68

 Initialized

0.64

Rim

0.63

 resourceCulture

0.61



0.61

 pond

0.58

 ioutil

0.55

 stole

0.54

ftagPool

0.54

Activations Density 0.204%

references to translation in various contexts

No Comments

No Known Activations