INDEX

Explanations

terms related to cataloging or listing items

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

ä¿Ĺ

-0.18

 weather

-0.15

zw

-0.15

weather

-0.14

 Weather

-0.14

ISCO

-0.14

 Äĳáº¡o

-0.14

lessly

-0.14

ollar

-0.14

 ÄĳÃ¢y

-0.14

POSITIVE LOGITS

ues

0.20

matic

0.19

ued

0.19

mium

0.18

ue

0.18

mate

0.17

strup

0.17

phere

0.17

aurus

0.17

ueue

0.16

Activations Density 0.012%

terms related to cataloging or listing items

No Comments

No Known Activations