INDEX

Explanations

numerical references and identifiers in texts

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

iggers

-0.21

menin

-0.14

aju

-0.14

ObjectContext

-0.14

RouterModule

-0.14

igg

-0.13

kee

-0.13

oron

-0.13

iasi

-0.13

ÑıÐ²Ð¸

-0.13

POSITIVE LOGITS

odash

0.16

Ð¾ÑĤÑĮ

0.16

annes

0.14

į°

0.14

 ÎłÏģÏī

0.14

ered

0.14

addle

0.14

:///

0.14

efe

0.13

leon

0.13

Activations Density 0.001%

numerical references and identifiers in texts

No Comments

No Known Activations