INDEX

Explanations

terms related to governmental or institutional authority and operations

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

serie

-0.17

egrator

-0.16

uristic

-0.15

antal

-0.15

wise

-0.15

æĻ

-0.14

ÑĸÐ½ÑĮ

-0.14

Ä±ÅŁÄ±k

-0.14

liste

-0.14

yth

-0.14

POSITIVE LOGITS

ningen

0.39

lingen

0.32

standen

0.31

ansen

0.29

heten

0.28

elsen

0.27

ingen

0.26

orna

0.25

isten

0.24

loven

0.24

Activations Density 0.060%

terms related to governmental or institutional authority and operations

No Comments

No Known Activations