INDEX

Explanations

references to characters in various forms of narratives

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

adpleegd

-0.57

 setLoading

-0.55

 Abbey

-0.53

OGND

-0.52

 Blitz

-0.52

 {}));

-0.52

Erreferentziak

-0.52

 <=",

-0.52

|}{}

-0.50

 Bisnis

-0.50

POSITIVE LOGITS

Character

0.94

 Character

0.92

 character

0.92

 Characters

0.87

character

0.87

 characters

0.84

Characters

0.83

characters

0.80

 CHARACTER

0.80

CHARACTER

0.78

Activations Density 0.109%

references to characters in various forms of narratives

No Comments

No Known Activations