INDEX

Explanations

references to existing methods or sources of information

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

 input

-0.46

ren

-0.45

Med

-0.45

NUMX

-0.42

 attribute

-0.42

Doc

-0.42

 Consist

-0.41

 Input

-0.41

Ƿ

-0.40

 attributes

-0.40

POSITIVE LOGITS

exhaust

0.71

 exhaust

0.64

Exhaust

0.63

 wastes

0.63

exhaus

0.57

 Савезне

0.57

waste

0.57

 Exhaust

0.57

 épu

0.57

 bestehende

0.55

Activations Density 0.364%

references to existing methods or sources of information

No Comments

No Known Activations

references to existing methods or sources of information

No Comments

No Known Activations