INDEX

Explanations

words or phrases related to location or position

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

IVEREF

-0.69

 noDo

-0.60

na

-0.60

mergeFrom

-0.59

 незавершена

-0.56

ViewImports

-0.54

expandindo

-0.54

IndentedString

-0.51

RegistryLite

-0.50

Obrázky

-0.50

POSITIVE LOGITS

 referenties

0.36



0.33

越好

0.32

 gärna

0.28

Hap

0.28

asteroide

0.27

vbo

0.27

Av

0.27

男性

0.26

ναν

0.26

Activations Density 0.001%

words or phrases related to location or position

No Comments

No Known Activations