INDEX

Explanations

structured information related to instructions and indicators

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

 synonym

-0.16

 adel

-0.16

 Twig

-0.14

msp

-0.14

 Shades

-0.14

ymes

-0.14

alta

-0.14

Pep

-0.14

elen

-0.14

adena

-0.14

POSITIVE LOGITS

æĺĵ

0.18

 signal

0.17

à¸ļà¸Ńà¸ģ

0.16

identity

0.16

 identity

0.16

OMEM

0.16

 Identity

0.16

orde

0.16

indic

0.16

èŃ¦

0.15

Activations Density 0.132%

structured information related to instructions and indicators

No Comments

No Known Activations