INDEX

Explanations

specific details and references related to publication history and historical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

#af

-0.17

amat

-0.15

aroo

-0.15

azzi

-0.14

ë³

-0.14

_reads

-0.14

artner

-0.14

 Comple

-0.14

 ÐºÐ»Ñĥ

-0.13

PIO

-0.13

POSITIVE LOGITS

izada

0.17

ivec

0.16

trak

0.15

Ãł

0.15

/or

0.14

izable

0.14

 origin

0.14

Activations Density 0.158%

specific details and references related to publication history and historical context

No Comments

No Known Activations