INDEX

Explanations

phrases indicating procedural contexts or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B @ 39-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_39/width_16k/average_l0_131

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.39.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

 lisäksi

-0.52

 betrekking

-0.51

 particulière

-0.49

 yeux

-0.47

 épaules

-0.47

 particulières

-0.47

StoreMessageInfo

-0.46

 PARTICULAR

-0.46

 particular

-0.45

 respecto

-0.44

POSITIVE LOGITS

 batches

0.73

 increments

0.71

 accordance

0.70

 stages

0.63

offen

0.60

 groups

0.60

 pairs

0.59

 secrecy

0.59

 phases

0.57

 harmony

0.57

Activations Density 0.771%

phrases indicating procedural contexts or conditions

No Comments

No Known Activations