INDEX

Explanations

specific medical imaging techniques and their results

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

 rele

-0.15

inker

-0.15

[System

-0.14

èª

-0.14

jem

-0.14

éħ¸

-0.14

 Coul

-0.14

dib

-0.13

::$

-0.13

å®¢

-0.13

POSITIVE LOGITS

 enhancement

0.26

 enhancing

0.25

 Enhancement

0.23

-enh

0.20

gad

0.20

Rim

0.20

 appearances

0.19

Gad

0.19

 enhance

0.19

Enh

0.19

Activations Density 0.011%

specific medical imaging techniques and their results

No Comments

No Known Activations