INDEX

Explanations

references to particular individuals or characters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 8-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_8/width_16k/average_l0_71

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.8.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

 sacré

-0.94

windowFixed

-0.84

ⓧ

-0.84

 noDo

-0.81

 DriverManager

-0.80

 africains

-0.80

 récomp

-0.78

AnchorStyles

-0.77

AsUp

-0.76

 olímp

-0.76

POSITIVE LOGITS

Ray

0.82

 Weller

0.77

łuż

0.76

RAY

0.73

RAY

0.71

 Holmes

0.70

Rai

0.69

Ray

0.68

</i>

0.67

Activations Density 1.993%

references to particular individuals or characters

No Comments

No Known Activations