INDEX

Explanations

references to specific individuals or groups and their affiliations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

alion

-0.16

_decode

-0.15

ystick

-0.15

 Canter

-0.15

atu

-0.15

Native

-0.14

attach

-0.14

ulling

-0.14

isko

-0.14

ProgressHUD

-0.14

POSITIVE LOGITS

 Gros

0.20

 Bunun

0.19

 ÑĤÐ²Ð°

0.17

 Bunu

0.17

 Karen

0.17

 Carrier

0.16

Carrier

0.16

 éĸ

0.16

 Cube

0.15

 Delaware

0.15

Activations Density 0.057%

references to specific individuals or groups and their affiliations

No Comments

No Known Activations