INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

£½

-0.15

assa

-0.14

leet

-0.14

kke

-0.14

ÑĭÑĪ

-0.14

phan

-0.13

Ã¦re

-0.13

poz

-0.13

inux

-0.13

atur

-0.13

POSITIVE LOGITS

abee

0.15

rosse

0.14

htm

0.14

awi

0.14

/Instruction

0.14

ierz

0.13

olik

0.13

amon

0.13

mue

0.13

ÈĽi

0.13

Activations Density 0.000%

No Known Activations

This feature has no known activations.