INDEX

Explanations

terms related to software configuration and options

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_131k/average_l0_114

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

 createSlice

-0.41

 مرئيه

-0.40

 manifest

-0.36

rzost

-0.36

 Audiodateien

-0.35

DockStyle

-0.35

NewGuid

-0.33

ARROLL

-0.33

来歴

-0.32

 pompa

-0.32

POSITIVE LOGITS

 betweenstory

0.77

 ModelExpression

0.60

IUrlHelper

0.54

WarningLevel

0.53

gonic

0.49

 argint

0.48

 ProtoMessage

0.47

 intptr

0.47

AutoField

0.47

aarrggbb

0.46

Activations Density 0.029%

terms related to software configuration and options

No Comments

No Known Activations