INDEX

Explanations

references to various types of arms

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

Referencies

-0.45

 AssemblyTitle

-0.42

apparence

-0.42

sizeCache

-0.41

ртка

-0.40

 kasarigan

-0.40

 surla

-0.39

astrous

-0.39

 lezen

-0.38

 Olsson

-0.38

POSITIVE LOGITS

arm

0.66

Arm

0.65

 frost

0.60

BeginContext

0.58

frost

0.58

Arm

0.57

arm

0.57

 Frost

0.56

Frost

0.51

 Arms

0.50

Activations Density 1.507%

references to various types of arms

No Comments

No Known Activations