INDEX

Explanations

phrases related to evidence, factors, and legal contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 40-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_40/width_16k/average_l0_125

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.40.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

 nonUne

-0.53

########.

-0.44

 Infórmanos

-0.44

 PopupWindow

-0.42

巯

-0.41

wiſe

-0.40

TextWatcher

-0.38

 noDo

-0.37

melidir

-0.36

zot

-0.36

POSITIVE LOGITS

 leaning

1.68

 favoring

1.65

 favour

1.60

 bias

1.59

 favor

1.59

 biased

1.57

 leans

1.56

 tilted

1.55

 skewed

1.51

 leaned

1.47

Activations Density 1.129%

phrases related to evidence, factors, and legal contexts

No Comments

No Known Activations