INDEX

Explanations

references to copyright and ownership details

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

.TXT

-0.16

WARDED

-0.16

ibel

-0.16

armor

-0.15

renom

-0.15

.bpm

-0.14

istring

-0.14

á¿¦

-0.14

_hierarchy

-0.13

Appending

-0.13

POSITIVE LOGITS

 design

0.23

 theme

0.23

 designed

0.22

 Site

0.22

powered

0.22

 privacy

0.21

 powered

0.21

 terms

0.21

 Designed

0.21

 Powered

0.21

Activations Density 0.078%

references to copyright and ownership details

No Comments

No Known Activations