INDEX

Explanations

phrases indicating remaining quantities or presence after reductions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

 acompan

-0.15

htdocs

-0.14

 Reform

-0.14

 Minds

-0.14

pri

-0.14

ulumi

-0.14

/libs

-0.13

coni

-0.13

rift

-0.13

.skills

-0.13

POSITIVE LOGITS

 Remaining

0.22

 remaining

0.22

remaining

0.21

Remaining

0.20

æ®ĭ

0.16

 Standing

0.16

ategorie

0.16

åī©

0.16

eworld

0.15

ÐµÑģÐ¿

0.15

Activations Density 0.179%

phrases indicating remaining quantities or presence after reductions

No Comments

No Known Activations