INDEX

Explanations

questions and inquiries about understanding and knowledge

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 27-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_27/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.27.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

 فريبيس

-0.54

IANGLES

-0.50

CodeAttribute

-0.48

arraycopy

-0.45

 actionMode

-0.44

 Moscú

-0.43

findpost

-0.42

 wikipagina

-0.40

 Divulgação

-0.40

RTDA

-0.40

POSITIVE LOGITS

 suddenly

0.47

 Италијани

0.43

get

0.42

 survived

0.42

 magically

0.42

 manage

0.41

 sudden

0.41

Życiorys

0.40

imp

0.40

Activations Density 0.126%

questions and inquiries about understanding and knowledge

No Comments

No Known Activations