INDEX

Explanations

phrases indicating value or worth 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 @ 0-gemmascope-res-16k

Configuration

google/gemma-scope-9b-pt-res/layer_0/width_16k/average_l0_129

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.0.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

 combination

-0.47

 windows

-0.45

ured

-0.45

 Schild

-0.45

 utilization

-0.44

 powered

-0.43

 thủy

-0.43

 full

-0.43

Hen

-0.43

 municipality

-0.42

POSITIVE LOGITS

 nakalista

0.74

DockStyle

0.72

MessageOf

0.71

 speranza

0.68

CloseOperation

0.68

bcryptjs

0.67

BagConstraints

0.66



0.65

ⓧ

0.65

 esternos

0.65

Activations Density 0.303%

phrases indicating value or worth in various contexts

No Comments

No Known Activations