INDEX

Explanations

references related to bibliographic and cataloging information

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

ubes

-0.15

_Impl

-0.15

 pots

-0.14

akest

-0.14

OrElse

-0.14

ront

-0.14

umpt

-0.14

-pill

-0.14

orthand

-0.14

gili

-0.14

POSITIVE LOGITS

rance

0.16

 Consort

0.15

 intermitt

0.14

YNAM

0.14

Yug

0.14

iston

0.13

ehir

0.13

itched

0.13

 safeg

0.13

isha

0.13

Activations Density 0.003%

references related to bibliographic and cataloging information

No Comments

No Known Activations