INDEX

Explanations

concepts and objects that share a common type or category

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

atest

-0.15

 Trot

-0.14

 Venice

-0.14

{name

-0.14

elles

-0.14

SCO

-0.14

.libs

-0.14

ebi

-0.14

/Admin

-0.14

bsd

-0.14

POSITIVE LOGITS

ãģĶ

0.18

Ð¼Ð¾Ð½

0.17

ily

0.16

 benzer

0.15

POS

0.15

ë¥ĺ

0.15

nock

0.14

ushima

0.14

 Peak

0.14

ones

0.14

Activations Density 0.073%

concepts and objects that share a common type or category

No Comments

No Known Activations