INDEX

Explanations

terms related to applicability and regulations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

 Tape

-0.16

tape

-0.16

 tape

-0.16

 invisible

-0.15

$__

-0.15

invisible

-0.15

uger

-0.14

Ã´n

-0.14

bai

-0.13

hdl

-0.13

POSITIVE LOGITS

eyJ

0.16

alth

0.16

="{!!

0.15

0.14

gua

0.14

ToWorld

0.14

èıĮ

0.14

ÏĢÎ¿Ïį

0.14

ë©´ìłģ

0.14

íĥģ

0.14

Activations Density 0.033%

terms related to applicability and regulations

No Comments

No Known Activations