INDEX

Explanations

specific keywords and elements related to programming, logging, and data structure

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GEMMA-2-2B @ 15-gemmascope-res-16k

Configuration

google/gemma-scope-2b-pt-res/layer_15/width_16k/average_l0_78

Prompts (Dashboard)

36,864 prompts, 128 tokens each

Dataset (Dashboard)

monology/pile-uncopyrighted

Features

16,384

Data Type

float32

Hook Name

blocks.15.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

fortawesome

-0.40

at

-0.40

fan

-0.40

At

-0.40

vän

-0.39

 està

-0.39

 того

-0.38

mstyle

-0.38

nap

-0.38

POSITIVE LOGITS

RegressionTest

0.96

 Савезне

0.96

etc

0.89

참고

0.85

Etc

0.82

省市镇

0.81

 myſelf

0.81

قایناقلار

0.80

什么的

0.78

 kaarangay

0.76

Activations Density 0.337%

specific keywords and elements related to programming, logging, and data structure

No Comments

No Known Activations