INDEX

Explanations

punctuation marks, specifically periods and quotation marks

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 jadx

-0.17

 Kaynak

-0.14

à§į

-0.14

 ÙĪØ§ÙĦØŃ

-0.14

~-~-~-~-

-0.14

sher

-0.14

Ð±ÑĥÐ´ÑĮ

-0.14

ÑĢÑĸÐ´

-0.13

alon

-0.13

ì¼ĵ

-0.13

POSITIVE LOGITS

 ÐĴ

0.20

ÐŁÐ¾ÑģÐ»Ðµ

0.19

ÐĴ

0.19

0.18

Po

0.18

ÐĿÐ°

0.17

 ÐŁÐ¾ÑģÐ»Ðµ

0.17

 Ð²

0.15

 ÐŁÐ¾

0.15

Activations Density 0.012%

punctuation marks, specifically periods and quotation marks

No Comments

No Known Activations