INDEX

Explanations

references to specific scientific terminology or notations

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

 vixion

-0.45

 kitchen

-0.38

 הק

-0.36

 kinetics

-0.36

jetunion

-0.35

 kier

-0.33

き

-0.32

chlag

-0.32

 argint

-0.31

 decorazione

-0.31

POSITIVE LOGITS

KK

1.59

KK

1.56

KP

1.37

KC

1.37

KT

1.34

KM

1.33

KP

1.33

Ｋ

1.31

KS

1.31

PK

1.30

Activations Density 0.742%

references to specific scientific terminology or notations

No Comments

No Known Activations