INDEX

Explanations

technical terms and keywords related to programming and data structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

iger

-0.16

 latter

-0.15

ãĥĺ

-0.15

ht

-0.15

getic

-0.14

Ã§Ã£o

-0.14

ë¨¸

-0.14

ÑģÑĤÐ¸

-0.14

ooth

-0.14

inium

-0.14

POSITIVE LOGITS

apolis

0.16

lÃ¡n

0.16

ettel

0.15

olini

0.15

Ä°Y

0.15

metic

0.15

 Reaper

0.15

asan

0.14

Ã£

0.14

ichen

0.14

Activations Density 0.082%

technical terms and keywords related to programming and data structures

No Comments

No Known Activations