INDEX

Explanations

segments of code or programming constructs

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

########.

-0.59

 autorytatywna

-0.58

 '\\;'

-0.54

Autoritní

-0.52

 patrio

-0.50

COUVER

-0.49

 pron

-0.47

tvguidetime

-0.47

CF

-0.46

ONTO

-0.46

POSITIVE LOGITS

UnknownFieldSet

0.54

 gezondheid

0.47

enumi

0.45

 <<<<<<<<<<<<<<

0.43

 děti

0.42

 pagină

0.42

AndEndTag

0.42

 ligiloj

0.41

 prefeitura

0.40

hygiène

0.39

Activations Density 0.123%

segments of code or programming constructs

No Comments

No Known Activations