INDEX

Explanations

references to equipment and preparations for training activities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

 Borders

-0.18

 borders

-0.17

odos

-0.16

ump

-0.15

 applied

-0.14

vem

-0.14

ador

-0.14

anger

-0.14

exh

-0.14

POSITIVE LOGITS

 nehmen

0.16

okable

0.15

 chiá»ģu

0.15

.contentSize

0.15

èĳ¡

0.14

 ØŃÙĪ

0.14

ãĥĥãĤ¯ãĤ¹

0.14

 Ð´ÑĥÑħ

0.14

hai

0.14

 Ð´Ð¸Ð°Ð¼ÐµÑĤ

0.14

Activations Density 0.059%

references to equipment and preparations for training activities

No Comments

No Known Activations