INDEX

Explanations

references and citations to academic journals, articles, and publications

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

undra

-0.16

 GlobalKey

-0.15

marvin

-0.15

coli

-0.15

olean

-0.15

ibu

-0.15

abant

-0.14

jeta

-0.14

rack

-0.14

inkel

-0.14

POSITIVE LOGITS

åĲ¾

0.16

æ¿

0.15

unw

0.15

archy

0.14

dl

0.14

bir

0.14

atism

0.14

ÏģÎ¹ÏĥÏĦ

0.14

 Weston

0.14

 main

0.14

Activations Density 0.014%

references and citations to academic journals, articles, and publications

No Comments

No Known Activations