INDEX

Explanations

code-related sections involving function definitions and return statements

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 21-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_21/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.21.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

intenant

-0.43

 defaultstate

-0.39

bootstrapcdn

-0.38

 refinements

-0.34

 appointments

-0.34

 discoveries

-0.33

 FACTS

-0.32

 instructions

-0.31

 slowest

-0.31

 selections

-0.30

POSITIVE LOGITS

 يتيمه

0.54

Билгалдахарш

0.54

 kasarigan

0.52

copg

0.52

 miniaturka

0.50

 ResponseEntity

0.50

RectangleBorder

0.49

 jadx

0.49

 eût

0.49

coroutines

0.49

Activations Density 0.119%

code-related sections involving function definitions and return statements

No Comments

No Known Activations