INDEX

Explanations

technical specifications, parameters, and functions related to programming and hardware

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_16k/average_l0_91

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

elemField

-0.66

OGND

-0.64

 nonUne

-0.55

 surla

-0.54

 wireType

-0.54

WriteTagHelper

-0.52

posedge

-0.51

ValueStyle

-0.49

Viel

-0.47

+#+#

-0.47

POSITIVE LOGITS

tinyos

0.38

AutoScaleMode

0.38

OuterClass

0.35

 oraciones

0.33

 GenerationType

0.33

 Roskov

0.33

 Schie

0.31

erski

0.30

当初

0.29

archer

0.29

Activations Density 1.023%

technical specifications, parameters, and functions related to programming and hardware

No Comments

No Known Activations