INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

rott

-0.15

engin

-0.14

bjerg

-0.14

Ð²ÑģÑı

-0.14

 Ãĸr

-0.14

 challenger

-0.13

inoa

-0.13

 èĹ¤

-0.13

GuidId

-0.13

Ìģ

-0.12

POSITIVE LOGITS

URITY

0.15

 seasonal

0.14

.persistent

0.14

tement

0.14

tering

0.13

prite

0.13

SelectionMode

0.13

 Chains

0.13

ulary

0.13

à¤Ĩà¤ª

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.