issues
search
kimborgen
/
falcon-llm
Apache License 2.0
1
stars
0
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Hfport
#16
kimborgen
closed
1 year ago
0
Hardcode paralell attention and multi-query
#15
kimborgen
closed
1 year ago
0
Falcon 7b and 40b differing attention layer
#14
kimborgen
closed
1 year ago
1
LoRA on fused kvq performance
#13
kimborgen
opened
1 year ago
0
Optimize AttentionRotary
#12
kimborgen
opened
1 year ago
0
Split rotary alibi
#11
kimborgen
closed
1 year ago
1
Not supported: output_attentions
#10
kimborgen
closed
1 year ago
1
Hardcode pararell attention/MLP
#9
kimborgen
closed
1 year ago
2
Split Alibi and Rotary models
#8
kimborgen
closed
1 year ago
1
Fine-tune with alibi
#7
kimborgen
opened
1 year ago
0
Missing final MLP
#6
kimborgen
opened
1 year ago
0
parallel attention/mlp speedup
#5
kimborgen
opened
1 year ago
0
Architecture diagram
#4
kimborgen
closed
1 year ago
1
LayerNorm returns wrong dtype
#3
kimborgen
closed
1 year ago
1
Experiment with alibi
#2
kimborgen
closed
1 year ago
2
Past key values
#1
kimborgen
closed
1 year ago
5