INDEX

Explanations

references to hierarchical positions or rankings, particularly emphasizing the concept of "top."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_1/width_16k/average_l0_102

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

BufferException

-0.63

Carriera

-0.60

참고

-0.60

toHave

-0.56

()?;

-0.54

RenderAtEndOf

-0.53

oos

-0.53

toDouble

-0.53

ngilizce

-0.51

edema

-0.51

POSITIVE LOGITS

top

1.81

 bottom

0.97

top

0.91

 redor

0.82

 toppen

0.80

 highest

0.80

 behalf

0.80

上位

0.78

 upper

0.76

bottom

0.76

Activations Density 0.144%

references to hierarchical positions or rankings, particularly emphasizing the concept of "top."

No Comments

No Known Activations