INDEX

Explanations

references to a specific entity or brand name, specifically "Fu" and "foo"

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/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.0.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

principalColumn

-0.51

 ucran

-0.44

 documental

-0.43

 Ucrania

-0.43

 mecánica

-0.42

 extremely

-0.42

 매우

-0.40

 parís

-0.40

 ukrain

-0.40

 Asunción

-0.40

POSITIVE LOGITS

 booth

0.71

foo

0.67

 resourceCulture

0.66

Fu

0.63

Foo

0.62

Fu

0.61

 Sans

0.61

 Hood

0.60

foo

0.59

ista

0.58

Activations Density 0.218%

references to a specific entity or brand name, specifically "Fu" and "foo"

No Comments

No Known Activations