INDEX

Explanations

No Explanations Found

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

iph

-0.16

rouw

-0.16

unas

-0.15

æ¦

-0.15

ÌĨ

-0.14

Ģ

-0.14

 Parsons

-0.14

tap

-0.14

usercontent

-0.13

ensibly

-0.13

POSITIVE LOGITS

-aos

0.15

ichten

0.14

apgolly

0.14

idon

0.14

ublic

0.14

oultry

0.13

osis

0.13

 baiser

0.13

Activations Density 0.075%

No Comments

No Known Activations