INDEX

Explanations

references to entities and their attributes, particularly focusing on nouns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

EDGE

-0.15

uggy

-0.14

 Erick

-0.13

Ã¡j

-0.13

rawn

-0.13

ationale

-0.13

ixer

-0.13

etter

-0.12

 noses

-0.12

¼

-0.12

POSITIVE LOGITS

 fÃ¸rste

0.18

 first

0.17

 eerste

0.16

 Ð¿ÐµÑĢÐ²Ð¾Ð¹

0.16

 originals

0.16

 ì²«

0.15

 primeira

0.15

 FIRST

0.15

ç¬¬ä¸Ģ

0.15

.getFirst

0.15

Activations Density 0.008%

references to entities and their attributes, particularly focusing on nouns

No Comments

No Known Activations