INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

sled

-0.15

_ABI

-0.15

ãĤ·ãĥ£ãĥ«

-0.14

ndo

-0.14

pii

-0.14

leftJoin

-0.14

 ëĬ

-0.14

elden

-0.14

Ø´Ø§Ø±

-0.13

Ð¸ÑĤÐ¾Ð²

-0.13

POSITIVE LOGITS

roys

0.16

utz

0.16

ovan

0.15

 Chim

0.15

oux

0.15

udios

0.14

holm

0.13

ibu

0.13

Dal

0.13

Ð²Ð¸ÑĤ

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.