INDEX

Explanations

phrases related to technology, systems, or services in the context of efficiency and coverage

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

 blame

-0.15

Ø¶ÙĬ

-0.15

adir

-0.15

ÐµÐºÐ¾

-0.14

ìŁģ

-0.14

ÑħÑĥ

-0.14

Ð¾ÑĤÑĢÐµÐ±

-0.13

izik

-0.13

phia

-0.13

POSITIVE LOGITS

non

0.29

Non

0.25

Non

0.25

NON

0.24

non

0.24

 Outside

0.24

 beyond

0.23

_non

0.23

Outside

0.23

éĿŀ

0.23

Activations Density 0.391%

phrases related to technology, systems, or services in the context of efficiency and coverage

No Comments

No Known Activations