INDEX

Explanations

code snippets related to data parsing and manipulation

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



-0.62

 createSlice

-0.45

 CreateTagHelper

-0.44

gató

-0.38

сылкі

-0.37

تقاوى

-0.37

 stampa

-0.36

ski

-0.36

HideFlags

-0.35

 estekak

-0.35

POSITIVE LOGITS

@[+][

0.61

 forState

0.48

 endblock

0.47

DockStyle

0.45

 AssemblyCulture

0.45

+#+

0.44

oczes

0.43

enteuer

0.43

 awakeFromNib

0.42

PickerController

0.41

Activations Density 1.243%

code snippets related to data parsing and manipulation

No Comments

No Known Activations