INDEX

Explanations

expressions related to morality and ethical behavior

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 14-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_14/width_16k/average_l0_84

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.14.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

 purpoſe

-0.65

 leaſt

-0.61

 myſelf

-0.60

 whoſe

-0.59

 Inſ

-0.58

ſelf

-0.57

 sauvages

-0.56

 leſs

-0.55

Sì

-0.55

 houſe

-0.55

POSITIVE LOGITS

is

0.85

 will

0.66

 must

0.64

can

0.61

has

0.61

 would

0.59

was

0.58

may

0.55

 might

0.54

isn

0.54

Activations Density 0.688%

expressions related to morality and ethical behavior

No Comments

No Known Activations