INDEX

Explanations

terms related to health and medical conditions in a different language

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

junction

-0.15

oco

-0.14

quirrel

-0.14

Ð»ÑİÐ´

-0.14

 Buckingham

-0.14

setError

-0.14

Roe

-0.13

 æ°ĳ

-0.13

éķĩ

-0.13

ChildIndex

-0.13

POSITIVE LOGITS

 club

0.20

 ÑĦÑĥÑĤ

0.19

 match

0.18

Lig

0.18

 klub

0.17

 football

0.17

 ÑģÐµÐ·

0.17

 ÐºÐ»ÑĥÐ±

0.17

 Rubin

0.17

lig

0.17

Activations Density 0.065%

terms related to health and medical conditions in a different language

No Comments

No Known Activations