INDEX

Explanations

phrases that suggest enhancement or decoration in various contexts

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

 Kaynak

-0.15

vais

-0.15

lad

-0.14

osm

-0.14

 indeed

-0.14

 druÅ¾

-0.14

istrovstvÃŃ

-0.14

-0.13

adero

-0.13

POSITIVE LOGITS

 otherwise

0.54

otherwise

0.47

 Otherwise

0.44

 OTHERWISE

0.42

Otherwise

0.39

 proceedings

0.35

 sonst

0.25

 jinak

0.23

 Proceedings

0.21

åĲ¦

0.20

Activations Density 0.084%

phrases that suggest enhancement or decoration in various contexts

No Comments

No Known Activations