INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 Fucking

-0.17

 fucking

-0.16

 Fuck

-0.15

 Witness

-0.15

 fucked

-0.15

.yahoo

-0.15

erge

-0.15

...

-0.14

 witness

-0.14

POSITIVE LOGITS

 listener

0.21

 listeners

0.21

 Listener

0.19

listener

0.19

 /↵

0.17

ãĥªãĤ¹

0.17

Listener

0.17

/;↵

0.16

Listeners

0.16

Activations Density 0.000%

No Known Activations

This feature has no known activations.