INDEX

Explanations

references to electronics and computer components

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 Tune

-0.15

 crossorigin

-0.15

mund

-0.14

 Batter

-0.14

iface

-0.14

 brun

-0.13

bolt

-0.13

anchors

-0.13

nad

-0.13

POSITIVE LOGITS

 gates

0.27

 logic

0.26

 Logic

0.25

 logical

0.25

AND

0.24

 gate

0.23

gate

0.23

logic

0.23

 Boolean

0.22

 Exclusive

0.22

Activations Density 0.024%

references to electronics and computer components

No Comments

No Known Activations