INDEX

Explanations

names or variations of names

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

 المعيارى

-0.81

PerformLayout

-0.80

RegressionTest

-0.71

 Савезне

-0.68

########.

-0.67

DockStyle

-0.65

contentLoaded

-0.63

TypedDataSet

-0.60

Personendaten

-0.59

ITECH

-0.59

POSITIVE LOGITS

ar

0.38

Ephe

0.34

 burbujas

0.31

プーン

0.30

beginTransaction

0.29

 curvas

0.29

 podle

0.29

Cur

0.29

 intensidad

0.28

za

0.28

Activations Density 0.010%

names or variations of names

No Comments

No Known Activations