vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

update README.md

Browse source
This commit is contained in:
xaedes 2023-09-01 16:04:08 +02:00
parent 6cbf55a64b
commit 7acb1241c6
No known key found for this signature in database
GPG key ID: 30030EDD817EA2B1
1 changed files with 1 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

5
examples/finetune/README.md
View file

@ -15,8 +15,7 @@ wget https://raw.githubusercontent.com/brunoklein99/deep-learning-notes/master/s
--train-data "shakespeare.txt" \ --train-data "shakespeare.txt" \
--save-every 10 \ --save-every 10 \
--threads 6 --adam-iter 30 --batch 4 --ctx 64 \ --threads 6 --adam-iter 30 --batch 4 --ctx 64 \
--use-checkpointing --use-alloc \ --use-checkpointing
--mem-lora 2 --mem-compute 1 --mem-compute0 20
# predict # predict
./bin/main -m open-llama-3b-v2-q8_0.gguf --lora lora-open-llama-3b-v2-q8_0-shakespeare-LATEST.bin ./bin/main -m open-llama-3b-v2-q8_0.gguf --lora lora-open-llama-3b-v2-q8_0-shakespeare-LATEST.bin
@ -28,8 +27,6 @@ The pattern "ITERATION" in the output filenames will be replaced with the iterat
Gradient checkpointing reduces the memory requirements by ~50% but increases the runtime. Gradient checkpointing reduces the memory requirements by ~50% but increases the runtime.
If you have enough RAM, you can make finetuning a bit faster by disabling checkpointing with `--no-checkpointing`. If you have enough RAM, you can make finetuning a bit faster by disabling checkpointing with `--no-checkpointing`.
To change the amount of memory for finetuning with memory allocator (`--use-alloc`, used by default), you can use `--mem-compute0 N` to specify the number of gigabytes.
The LORA rank is configured for each model tensor type separately with these command line options: The LORA rank is configured for each model tensor type separately with these command line options:
```bash ```bash