issues
search
danbraunai
/
simple_stories_train
Trains small LMs. Designed for training on SimpleStories
3
stars
1
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Save model checkpoints to Wandb
#13
ryanhoangt
opened
1 week ago
0
Add a simple Wandb logging integration
#12
ryanhoangt
closed
1 week ago
0
First attempt at checkpoints
#11
ThomasWMarshall
closed
2 months ago
3
Wrote simple streaming dataloader
#10
lennart-finke
closed
1 week ago
8
Add Typing to Llama Training
#9
lennart-finke
closed
2 weeks ago
2
Allow for training on custom dataset and tokenizer
#8
danbraunai
opened
2 months ago
1
Log runs with wandb
#7
danbraunai
opened
2 months ago
0
Verify RMS norm
#6
danbraunai
opened
2 months ago
3
Make it easy to port model to transformerlens
#5
danbraunai
opened
2 months ago
1
Add typing to train_llama.py
#4
danbraunai
closed
2 weeks ago
0
Add checkpoints throughout training
#3
danbraunai
closed
2 months ago
0
Check that our implementation matches a real llama implementation
#2
danbraunai
opened
2 months ago
0
Add license information
#1
lennart-finke
closed
2 months ago
0