INDEX

Explanations

phrases related to updates and changes in information

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

apel

-0.16

ãĥ¼ãĥŀ

-0.15

ëĭĪìķĦ

-0.15

 vÅ¯bec

-0.14

æ³ī

-0.14

reu

-0.14

åİŁæĿ¥

-0.14

RING

-0.14

tec

-0.14

POSITIVE LOGITS

 WILL

0.15

 shortly

0.15

ati

0.15

uw

0.15

 soon

0.14

auss

0.14

orph

0.14

 will

0.14

will

0.14

ProgressHUD

0.13

Activations Density 0.231%

phrases related to updates and changes in information

No Comments

No Known Activations