INDEX

Explanations

terms related to detail and summary information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 37-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_37/width_16k/average_l0_124

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.37.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

 الرياضيه

-0.67

 <<<<<<<<<<<<<<

-0.56

httphttps

-0.50

copg

-0.49

jewództ

-0.47

ⓧ

-0.46

 виправивши

-0.46

 defaultstate

-0.43

saraba

-0.42

 Normdatei

-0.41

POSITIVE LOGITS

 detail

0.57

hamento

0.53

 Detail

0.53

RenderAtEndOf

0.52

 details

0.51

 particulars

0.51

 minuti

0.50

 Details

0.47

细节

0.44

 Particulars

0.44

Activations Density 0.146%

terms related to detail and summary information

No Comments

No Known Activations