INDEX

Explanations

references to neural network structures and functions

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

New Auto-Interp

Configuration

google/gemma-scope-9b-pt-mlp/layer_6/width_131k/average_l0_72

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

Data Type

float32

Hook Name

blocks.6.hook_mlp_out

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

 autorytatywna

-0.66

 '\\;'

-0.59

rungsseite

-0.56

 nakalista

-0.55

saraba

-0.53

 CreateTagHelper

-0.53

UserScript

-0.47

 виправивши

-0.44

 AppCompat

-0.44

Bioaccumulative

-0.43

POSITIVE LOGITS

ParallelGroup

0.42

 joaat

0.41

ölkerung

0.40

 Suiza

0.35

 Sommer

0.35

 femen

0.35

Пита

0.35

 iParam

0.35

Organisateur

0.34

nels

0.34

Activations Density 0.166%

references to neural network structures and functions

No Comments

No Known Activations