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

fuck

-0.16

 Dude

-0.15

 fucking

-0.15

 dude

-0.15

 Fucking

-0.14

 shit

-0.14

 fucks

-0.14

 subsequent

-0.14

adesh

-0.14

 subsequently

-0.14

POSITIVE LOGITS

mo

0.19

mo

0.18

bus

0.18

 happy

0.17

 happiness

0.17

 Trom

0.16

 trom

0.16

.Bus

0.16

 sill

0.16

happy

0.16

Activations Density 0.000%

No Known Activations

This feature has no known activations.