INDEX

Explanations

references to competition and awards

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

znik

-0.16

cheng

-0.16

kol

-0.15

ãģĭãģĦ

-0.15

errer

-0.15

icari

-0.15

/gtest

-0.14

illez

-0.14

acco

-0.14

idar

-0.14

POSITIVE LOGITS

 Overall

0.15

 shorten

0.15

 éĻ

0.14

ãĥ¼ãĥ³

0.14

ghan

0.13

LOBAL

0.13

Pav

0.13

ä½ķ

0.13

omon

0.13

unction

0.13

Activations Density 0.039%

references to competition and awards

No Comments

No Known Activations