INDEX

Explanations

references to data types, particularly byte-related constructs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 7-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_7/width_16k/average_l0_69

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.7.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

AsUp

-0.66

 Ruskin

-0.66

opsida

-0.65

 Ελλά

-0.63

ized

-0.59

 film

-0.57

 Cher

-0.56

in

-0.54

 recupero

-0.53

 ffilm

-0.53

POSITIVE LOGITS

byte

1.80

 byte

1.51

Byte

1.12

 purpoſe

0.99

 BYTE

0.97

 Byte

0.96

 ſtand

0.95

 ſtate

0.95

 myſelf

0.91

BYTE

0.91

Activations Density 0.025%

references to data types, particularly byte-related constructs

No Comments

No Known Activations