INDEX

Explanations

references to research studies and academic writings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 38-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_38/width_16k/average_l0_128

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.38.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

 informée

-0.43

 Glaubens

-0.39

 optimizing

-0.39

 referenties

-0.37

 optimization

-0.37

 loup

-0.37

 optimisation

-0.37

 décid

-0.37

optimize

-0.36

 betweenstory

-0.36

POSITIVE LOGITS

 examine

1.25

 explore

1.15

 examines

1.13

 examined

1.08

 examining

1.06

 explores

1.02

examine

1.02

 explored

1.01

 Examine

1.00

Examine

1.00

Activations Density 0.559%

references to research studies and academic writings

No Comments

No Known Activations