INDEX

Explanations

phrases indicating difficulty or complexity in understanding issues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

 punt

-0.15

vel

-0.14

<Service

-0.14

ãĥ³ãĤ¯

-0.14

Äĵ

-0.14

 Booth

-0.14

udu

-0.14

kb

-0.14

 contra

-0.13

ware

-0.13

POSITIVE LOGITS

 simple

0.23

 simples

0.20

 simplest

0.18

 basic

0.18

 Simple

0.18

simple

0.17

 straightforward

0.17

andel

0.17

/basic

0.17

-simple

0.16

Activations Density 0.145%

phrases indicating difficulty or complexity in understanding issues

No Comments

No Known Activations