INDEX

Explanations

references to loudness or noise in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

google/gemma-scope-9b-pt-res/layer_2/width_16k/average_l0_67

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.2.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

rungsseite

-0.55

 bVar

-0.47

orited

-0.46

Behaviors

-0.45

tslib

-0.44

имущества

-0.43

libft

-0.43



-0.42

 seuil

-0.42

хьтан

-0.41

POSITIVE LOGITS

 loud

1.07

loud

0.92

Loud

0.81

 Loud

0.78

 population

0.71

 loudest

0.69

 louder

0.67

 loudly

0.63

population

0.63

 Population

0.59

Activations Density 0.065%

references to loudness or noise in various contexts

No Comments

No Known Activations