INDEX

Explanations

numerical data and historical references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

isper

-0.16

era

-0.16

@student

-0.15

abil

-0.15

ãĥ¯ãĥ¼

-0.15

ags

-0.15

apol

-0.15

/^(

-0.14

 epoch

-0.14

 authDomain

-0.14

POSITIVE LOGITS

baum

0.15

ÑĥÐ¼Ñĥ

0.14

$MESS

0.14

 Territory

0.14

Ã¸j

0.14

jez

0.14

XHR

0.14

Ã©ment

0.13

riet

0.13

Activations Density 0.015%

numerical data and historical references

No Comments

No Known Activations