INDEX

Explanations

numerical values and concepts related to statistics or performance metrics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

ÏĮ

-0.15

aber

-0.15

aspers

-0.14

phan

-0.14

 premium

-0.14

/xhtml

-0.14

¤ĳ

-0.14

tera

-0.13

à¸±à¸ĩ

-0.13

POSITIVE LOGITS

 advanced

0.21

 advancement

0.20

 interference

0.18

RBI

0.18

 advancements

0.18

LOB

0.17

 Team

0.17

 advances

0.17

advanced

0.17

ople

0.17

Activations Density 0.000%

numerical values and concepts related to statistics or performance metrics

No Comments

No Known Activations