INDEX

Explanations

references to specific publications or editions of books

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

abo

-0.17

hazi

-0.17

 everyone

-0.16

achi

-0.16

-0.15

ree

-0.14

 Nicholas

-0.14

pot

-0.14

POSITIVE LOGITS

angkan

0.16

kro

0.16

ÙħØ§Ùħ

0.15

apg

0.15

.Foundation

0.15

yonel

0.15

.fm

0.15

 suce

0.14

PointSize

0.14

stru

0.14

Activations Density 0.920%

references to specific publications or editions of books

No Comments

No Known Activations