INDEX

Explanations

punctuation marks in text

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

 disambiguazione

-0.61

+#+#

-0.54

⎠

-0.51

specifier

-0.48

UpInside

-0.45

 betweenstory

-0.44

openqa

-0.44

 gober

-0.44

 femininas

-0.44

reactstrap

-0.43

POSITIVE LOGITS

цездатний

0.40

boarding

0.35

flashdata

0.33

 parseFrom

0.32

 Inters

0.32

 Toss

0.31

banding

0.29

bitField

0.29

cur

0.29

AlterField

0.29

Activations Density 0.060%

punctuation marks in text

No Comments

No Known Activations