INDEX

Explanations

conversational elements and personal reflections in discussions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

Ã¡p

-0.15

 strained

-0.15

ãĥ¼ãĥ³

-0.14

okus

-0.14

lero

-0.14

zos

-0.13

-corner

-0.13

ÐµÐ´ÑĮ

-0.13

sdale

-0.13

 Uncategorized

-0.12

POSITIVE LOGITS

quez

0.15

dra

0.15

hi

0.13

ikat

0.13

^n

0.13

BarItem

0.13

 âĢ¦

0.13

bai

0.13

oooooooo

0.13

 MatSnackBar

0.13

Activations Density 0.439%

conversational elements and personal reflections in discussions

No Comments

No Known Activations