INDEX

Explanations

punctuation marks and sentence structures that indicate emotion or surprise

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

 lately

-0.17

azel

-0.15

 currently

-0.14

 recent

-0.14

 hopefully

-0.14

 particularly

-0.14

ffen

-0.14

errated

-0.14

currently

-0.13

 Ð´Ð¾Ðº

-0.13

POSITIVE LOGITS

 turns

0.27

 Turns

0.26

 immediately

0.25

Turn

0.22

 instantly

0.22

 immedi

0.21

 turned

0.21

 shortly

0.21

 Turn

0.20

 Needless

0.20

Activations Density 0.214%

punctuation marks and sentence structures that indicate emotion or surprise

No Comments

No Known Activations