INDEX

Explanations

documentation comments in a programming context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 11-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_11/width_16k/average_l0_118

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.11.hook_resid_post

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

 للمعارف

-0.57

wegg

-0.55

ніципалі

-0.54

 Wikimedijinoj

-0.54

CROSSTALK

-0.54

鰭

-0.54

Rujuakan

-0.53

𝕫

-0.53

PerformLayout

-0.53

–

-0.50

POSITIVE LOGITS

/**

2.14

/**

1.70

/**

1.53

/**

1.43

///**

1.30

+/**

1.23

/**

1.04

/**

0.97

-/**

0.97

/***

0.93

Activations Density 0.087%

documentation comments in a programming context

No Comments

No Known Activations