INDEX

Explanations

acronyms, particularly in scientific contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

OAD

-0.16

lyph

-0.16

ÑīÐ¸

-0.15

ippers

-0.14

amiento

-0.14

wie

-0.14

andes

-0.14

Ã¼p

-0.14

ornings

-0.14

olec

-0.14

POSITIVE LOGITS

atto

0.16

 Atkins

0.15

pek

0.15

uls

0.15

.BorderFactory

0.14

chester

0.13

 Ø§Ø¶

0.13

ced

0.13

ëĵľë¦¬

0.13

Activations Density 0.071%

acronyms, particularly in scientific contexts

No Comments

No Known Activations