INDEX

Explanations

terms related to procedural or instructional contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

OGND

-0.57

 Administrativna

-0.55

ftagPool

-0.54

 оригіналу

-0.54

 CanadaChoose

-0.53

 <=",

-0.52

хьтан

-0.51

 Biôgrafia

-0.51

 sumpay

-0.50

@[+][

-0.49

POSITIVE LOGITS

+:+

0.52

 however

0.45

ent

0.40

 &___

0.39

However

0.39

Jos

0.38

 aber

0.37

My

0.37

Jos

0.37

 also

0.36

Activations Density 0.752%

terms related to procedural or instructional contexts

No Comments

No Known Activations