INDEX

Explanations

terms related to formal categorization and procedure

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

Ã£

-0.17

icans

-0.16

CUS

-0.15

pit

-0.15

ãĤ¤ãĥ¤

-0.14

avigate

-0.14

 ÑģÐ¾Ð±Ð¾Ñİ

-0.14

iale

-0.14

ibase

-0.14

 Ð·Ð°Ð²Ð¸

-0.14

POSITIVE LOGITS

 means

0.26

 virtue

0.22

means

0.20

 analogy

0.20

 standards

0.19

 proxy

0.18

 Proxy

0.17

 alone

0.17

 Means

0.16

Proxy

0.15

Activations Density 0.166%

terms related to formal categorization and procedure

No Comments

No Known Activations