INDEX

Explanations

technical terms and syntax related to programming and data handling

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 1-gemmascope-mlp-16k

Configuration

google/gemma-scope-2b-pt-mlp/layer_1/width_16k/average_l0_105

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.1.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

FXML

-0.59

 rehearing

-0.56

 pulito

-0.56

Xna

-0.53

coj

-0.53

djangoproject

-0.52

didSet

-0.52

 honour

-0.50

ainville

-0.50

новништво

-0.49

POSITIVE LOGITS

devtools

0.63

protoimpl

0.57

oplan

0.56

 transfieras

0.54

IndentedString

0.52

 entende

0.52

 AssemblyProduct

0.52

deelte

0.50

"%

0.50

 HasFactory

0.49

Activations Density 0.497%

technical terms and syntax related to programming and data handling

No Comments

No Known Activations