INDEX

Explanations

numerical information and statistics related to customer and employee figures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

lah

-0.16

czy

-0.14

reur

-0.14

/Dk

-0.14

_iface

-0.14

idian

-0.13

iquer

-0.13

agne

-0.13

quences

-0.13

erna

-0.13

POSITIVE LOGITS

etc

0.19

etc

0.15

 These

0.14

---

0.13

ottle

0.13

ãĥ©ãĥ³ãĥī

0.13

vs

0.13

com

0.13

Activations Density 0.092%

numerical information and statistics related to customer and employee figures

No Comments

No Known Activations