INDEX

Explanations

a variety of specific numerical identifiers and protocols typically associated with coding or data formats

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 36-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_36/width_16k/average_l0_120

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.36.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

 DataTypes

-0.41

ことができました

-0.40

sidemargin

-0.39

MemoryWarning

-0.39

 ویکی‌پدیای

-0.38

 maxim

-0.37

 uLocal

-0.36

協会

-0.35

 concret

-0.35

 harbor

-0.34

POSITIVE LOGITS

&-

1.04

*-

0.93

'-

0.87

based

0.81

®-

0.79

related

0.78

=-=-=-=-

0.76

»-

0.76

?-

0.75

'+

0.74

Activations Density 1.039%

a variety of specific numerical identifiers and protocols typically associated with coding or data formats

No Comments

No Known Activations