INDEX

Explanations

structured data marked up in XML format

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

Ã¢ce

-0.15

aghan

-0.15

 glyphicon

-0.14

ÐºÑĥÐ»ÑĮ

-0.14

Ð»Ð°Ð´Ñĥ

-0.14

pha

-0.14

quet

-0.14

istor

-0.13

Ð±Ð°Ð²

-0.13

hetics

-0.13

POSITIVE LOGITS

root

0.20

 root

0.17

 xmlns

0.17

RootElement

0.17

-root

0.16

 Root

0.16

ROOT

0.16

IDO

0.15

XML

0.15

Root

0.15

Activations Density 0.086%

structured data marked up in XML format

No Comments

No Known Activations