INDEX

Explanations

specific quantitative details related to measurement and sensitivity in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/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.21.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

 plutôt

-0.39

oneofs

-0.38

 ujednoznacz

-0.37

createClass

-0.37

 povezave

-0.37

 pratique

-0.36

يميديا

-0.36

 kuns

-0.36

MethodManager

-0.35

TagHelpers

-0.34

POSITIVE LOGITS

 slightest

0.82

 ANYTHING

0.56

 tiny

0.56

iniest

0.53

 Strict

0.52

ANY

0.51

 Anything

0.49

 smallest

0.48

期刊论文

0.47

anything

0.47

Activations Density 0.381%

specific quantitative details related to measurement and sensitivity in various contexts

No Comments

No Known Activations