INDEX

Explanations

specific numerical codes or symbols indicating various concepts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

ayd

-0.16

ROWS

-0.16

 AssemblyVersion

-0.16

à¸ļà¸£à¸£

-0.16

ills

-0.16

à¹ģà¸ļ

-0.15

ãĥ³ãĥĩ

-0.15

oger

-0.15

RIPT

-0.15

Ã¡ng

-0.15

POSITIVE LOGITS

 Tick

0.19

tun

0.17

 Tunnel

0.17

 Conf

0.16

 tunnel

0.16

Tun

0.16

Cop

0.16

til

0.16

Dlg

0.16

Tick

0.16

Activations Density 0.007%

specific numerical codes or symbols indicating various concepts

No Comments

No Known Activations