INDEX

Explanations

technical references to errors or malfunctions

oai_token-act-pair · gpt-4o-mini

Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 13-gemmascope-res-65k

Configuration

google/gemma-scope-2b-pt-res/layer_13/width_65k/average_l0_75

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

65,536

Data Type

float32

Hook Name

blocks.13.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

monology/pile-uncopyrighted

Activation Function

relu

Embeds

IFrame

Link

Negative Logits

Oleh

-0.96

gonic

-0.94

expandindo

-0.90

 producteurs

-0.88

 marié

-0.87

 فريبيس

-0.87

 culturali

-0.87

 MacGregor

-0.86

 venuto

-0.86

 maschile

-0.85

POSITIVE LOGITS

 fault

1.71

 Fault

1.50

fault

1.46

 faults

1.39

Fault

1.30

FAULT

1.23

faults

1.18

Faults

1.13

Fa

0.87

fau

0.84

Activations Density 0.005%

technical references to errors or malfunctions

No Known Activations