INDEX

Explanations

programming-related structures and commands

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

 Phong

-0.16

ering

-0.16

ille

-0.16

Pen

-0.15

ÙĥØª

-0.15

pen

-0.15

igen

-0.15

Dag

-0.14

 Paperback

-0.14

POSITIVE LOGITS

ÑĥÐ½Ðº

0.15

 flush

0.15

ahlen

0.15

 otherwise

0.15

ipel

0.14

Ð½Ð¸Ñĩ

0.14

Collider

0.14

.lucene

0.14

otherwise

0.14

 zÃ¡p

0.14

Activations Density 0.299%

programming-related structures and commands

No Comments

No Known Activations