INDEX

Explanations

keywords related to references and citations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

ester

-0.16

ibar

-0.15

ifu

-0.14

ernetes

-0.14

Ã¹ng

-0.14

onest

-0.14

abase

-0.14

pectrum

-0.14

utos

-0.14

Fur

-0.14

POSITIVE LOGITS

 Crown

0.17

hk

0.16

heimer

0.15

 Markus

0.15

Doe

0.15

 Revel

0.15

 Reyn

0.15

ounty

0.14

 reel

0.14

hi

0.14

Activations Density 0.023%

keywords related to references and citations

No Comments

No Known Activations