INDEX

Explanations

No Explanations Found

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

itu

-0.17

atcher

-0.16

epad

-0.16

eworld

-0.16

iglia

-0.15

ickey

-0.15

 pimp

-0.15

Ø±Ùī

-0.14

ames

-0.14

 crowd

-0.14

POSITIVE LOGITS

 Jeep

0.16

 jeep

0.16

oto

0.15

 goodness

0.15

 Humb

0.14

 until

0.14

 prepar

0.14

Gaw

0.14

çĽ´

0.14

pes

0.14

Activations Density 0.000%

No Known Activations

This feature has no known activations.