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

-0.16

abol

-0.15

Moreover

-0.14

WA

-0.14

resh

-0.14

ocus

-0.14

 ZÃ¡p

-0.14

ÙĪØ²

-0.14

Sept

-0.14

otu

-0.13

POSITIVE LOGITS

erras

0.19

 permitting

0.17

inus

0.17

inou

0.16

áº¹n

0.16

inu

0.14

/hash

0.14

oins

0.14

 likely

0.14

 amongst

0.14

Activations Density 0.000%

No Known Activations

This feature has no known activations.