INDEX

Explanations

specific planning or outlining formats

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 20-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_20/width_16k/average_l0_68

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.20.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

 desmotivaciones

-0.76

 Савезне

-0.73

 Infór

-0.65

 increí

-0.63

 estadounid

-0.63

 ſich

-0.63

 CanadaChoose

-0.62

 Geſch

-0.62

 ſes

-0.60

 ſever

-0.60

POSITIVE LOGITS

0.47

0.44

0.43

0.41

<--

0.41

-->

0.40

-->

0.40

ta

0.40

 sucks

0.39

os

0.39

Activations Density 0.154%

specific planning or outlining formats

No Comments

No Known Activations