INDEX

Explanations

terms related to medical treatments and procedures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

 autopsy

-0.15

 Basic

-0.15

 ÑĥÑĢÐ¾Ðº

-0.15

 Pact

-0.14

Basic

-0.14

 é¤

-0.14

é¤¨

-0.14

 Deng

-0.13

 clinically

-0.13

POSITIVE LOGITS

procedure

0.21

 treatment

0.20

 procedure

0.20

 candidacy

0.18

 technique

0.17

Performed

0.17

 treating

0.17

 Procedure

0.17

 treatments

0.17

 scar

0.17

Activations Density 0.099%

terms related to medical treatments and procedures

No Comments

No Known Activations