INDEX

Explanations

references to newly discovered species

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

 cuff

-0.16

EEP

-0.15

 laure

-0.14

enant

-0.14

èĬ¯

-0.14

Angel

-0.14

 ÑģÐ¾Ñģ

-0.14

Spl

-0.14

indi

-0.13

POSITIVE LOGITS

 spiders

0.39

 spider

0.37

 Spider

0.36

Spider

0.30

èĽĽ

0.30

èľĺèĽĽ

0.28

 webs

0.25

web

0.24

 Spinner

0.23

-web

0.22

Activations Density 0.013%

references to newly discovered species

No Comments

No Known Activations