INDEX

Explanations

actions or processes related to activation and functionality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-9B-IT @ 20-gemmascope-res-131k

Configuration

google/gemma-scope-9b-it-res/layer_20/width_131k/average_l0_81

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

131,072

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

del

-0.37

-0.35

was

-0.33

-0.31

ha

-0.31

↵↵

-0.31

POSITIVE LOGITS

 canst

0.87

 ويمكن

0.77

Puedes

0.75

ebilir

0.72

Podemos

0.72

Puede

0.71

 possiamo

0.71

 Pueden

0.71

verwijspagina

0.71

也可以

0.71

Activations Density 0.434%

actions or processes related to activation and functionality

No Comments

No Known Activations