consolidate as peft_model_dir

2023-09-19 19:02:14 -04:00
parent ba85308720
commit 203369411e
33 changed files with 50 additions and 49 deletions
--- a/examples/cerebras/btlm-ft.yml
+++ b/examples/cerebras/btlm-ft.yml
@@ -18,7 +18,7 @@ dataset_prepared_path: last_prepared_run
 val_set_size: 0.01

 adapter:
-lora_model_dir:
+peft_model_dir:
 sequence_len: 2048
 max_packed_sequence_len:
 sample_packing: false
--- a/examples/cerebras/qlora.yml
+++ b/examples/cerebras/qlora.yml
@@ -10,7 +10,7 @@ datasets:
 dataset_prepared_path:
 val_set_size: 0.01
 adapter: qlora
-lora_model_dir:
+peft_model_dir:
 sequence_len: 2048
 max_packed_sequence_len: 2048
 lora_r: 16
--- a/examples/code-llama/13b/lora.yml
+++ b/examples/code-llama/13b/lora.yml
@@ -20,7 +20,7 @@ sample_packing: true
 pad_to_sequence_len: true

 adapter: lora
-lora_model_dir:
+peft_model_dir:
 lora_r: 32
 lora_alpha: 16
 lora_dropout: 0.05
--- a/examples/code-llama/13b/qlora.yml
+++ b/examples/code-llama/13b/qlora.yml
@@ -16,7 +16,7 @@ val_set_size: 0.01
 output_dir: ./qlora-out

 adapter: qlora
-lora_model_dir:
+peft_model_dir:

 sequence_len: 4096
 sample_packing: true
--- a/examples/code-llama/34b/lora.yml
+++ b/examples/code-llama/34b/lora.yml
@@ -20,7 +20,7 @@ sample_packing: true
 pad_to_sequence_len: true

 adapter: lora
-lora_model_dir:
+peft_model_dir:
 lora_r: 32
 lora_alpha: 16
 lora_dropout: 0.05
--- a/examples/code-llama/34b/qlora.yml
+++ b/examples/code-llama/34b/qlora.yml
@@ -16,7 +16,7 @@ val_set_size: 0.01
 output_dir: ./qlora-out

 adapter: qlora
-lora_model_dir:
+peft_model_dir:

 sequence_len: 4096
 sample_packing: true
--- a/examples/code-llama/7b/lora.yml
+++ b/examples/code-llama/7b/lora.yml
@@ -20,7 +20,7 @@ sample_packing: true
 pad_to_sequence_len: true

 adapter: lora
-lora_model_dir:
+peft_model_dir:
 lora_r: 32
 lora_alpha: 16
 lora_dropout: 0.05
--- a/examples/code-llama/7b/qlora.yml
+++ b/examples/code-llama/7b/qlora.yml
@@ -16,7 +16,7 @@ val_set_size: 0.01
 output_dir: ./qlora-out

 adapter: qlora
-lora_model_dir:
+peft_model_dir:

 sequence_len: 4096
 sample_packing: true
--- a/examples/falcon/config-7b-lora.yml
+++ b/examples/falcon/config-7b-lora.yml
@@ -15,7 +15,7 @@ datasets:
 dataset_prepared_path:
 val_set_size: 0.01
 adapter: lora
-lora_model_dir:
+peft_model_dir:
 sequence_len: 2048
 max_packed_sequence_len:
 lora_r: 16
--- a/examples/falcon/config-7b-qlora.yml
+++ b/examples/falcon/config-7b-qlora.yml
@@ -22,7 +22,7 @@ dataset_prepared_path:
 val_set_size: 0.01
 # enable QLoRA
 adapter: qlora
-lora_model_dir:
+peft_model_dir:
 sequence_len: 2048
 max_packed_sequence_len:

--- a/examples/falcon/config-7b.yml
+++ b/examples/falcon/config-7b.yml
@@ -15,7 +15,7 @@ datasets:
 dataset_prepared_path:
 val_set_size: 0.01
 adapter:
-lora_model_dir:
+peft_model_dir:
 sequence_len: 2048
 max_packed_sequence_len:
 lora_r: 64
--- a/examples/gptj/qlora.yml
+++ b/examples/gptj/qlora.yml
@@ -10,7 +10,7 @@ datasets:
 dataset_prepared_path:
 val_set_size: 0.01
 adapter: qlora
-lora_model_dir:
+peft_model_dir:
 sequence_len: 2048
 max_packed_sequence_len:
 lora_r: 8
--- a/examples/jeopardy-bot/config.yml
+++ b/examples/jeopardy-bot/config.yml
@@ -9,7 +9,7 @@ datasets:
 dataset_prepared_path:
 val_set_size: 0.02
 adapter:
-lora_model_dir:
+peft_model_dir:
 sequence_len: 512
 max_packed_sequence_len:
 lora_r:
--- a/examples/llama-2/gptq-lora.yml
+++ b/examples/llama-2/gptq-lora.yml
@@ -18,7 +18,7 @@ datasets:
 dataset_prepared_path:
 val_set_size: 0.01
 adapter: lora
-lora_model_dir:
+peft_model_dir:
 sequence_len: 4096
 sample_packing:
 lora_r: 8
--- a/examples/llama-2/ia3.yml
+++ b/examples/llama-2/ia3.yml
@@ -20,7 +20,7 @@ sample_packing: true
 pad_to_sequence_len: true

 adapter: ia3
-ia3_model_dir:
+peft_model_dir:
 ia3_target_modules:
  - k_proj
  - v_proj
--- a/examples/llama-2/lora.yml
+++ b/examples/llama-2/lora.yml
@@ -20,7 +20,7 @@ sample_packing: true
 pad_to_sequence_len: true

 adapter: lora
-lora_model_dir:
+peft_model_dir:
 lora_r: 32
 lora_alpha: 16
 lora_dropout: 0.05
--- a/examples/llama-2/qlora.yml
+++ b/examples/llama-2/qlora.yml
@@ -16,7 +16,7 @@ val_set_size: 0.01
 output_dir: ./qlora-out

 adapter: qlora
-lora_model_dir:
+peft_model_dir:

 sequence_len: 4096
 sample_packing: true
--- a/examples/llama-2/relora.yml
+++ b/examples/llama-2/relora.yml
@@ -16,7 +16,7 @@ val_set_size: 0.01
 output_dir: ./relora-out

 adapter: qlora
-lora_model_dir:
+peft_model_dir:

 sequence_len: 4096
 sample_packing: true
--- a/examples/llama-2/tiny-llama.yml
+++ b/examples/llama-2/tiny-llama.yml
@@ -20,7 +20,7 @@ sequence_len: 4096
 sample_packing: true

 adapter: lora
-lora_model_dir:
+peft_model_dir:
 lora_r: 32
 lora_alpha: 16
 lora_dropout: 0.05
--- a/examples/mpt-7b/config.yml
+++ b/examples/mpt-7b/config.yml
@@ -9,7 +9,7 @@ datasets:
 dataset_prepared_path:
 val_set_size: 0.02
 adapter:
-lora_model_dir:
+peft_model_dir:
 sequence_len: 2048
 max_packed_sequence_len:
 lora_r: 8
--- a/examples/openllama-3b/config.yml
+++ b/examples/openllama-3b/config.yml
@@ -12,7 +12,7 @@ datasets:
 dataset_prepared_path:
 val_set_size: 0.02
 adapter:
-lora_model_dir:
+peft_model_dir:
 sequence_len: 1024
 sample_packing: true
 lora_r:
--- a/examples/openllama-3b/lora.yml
+++ b/examples/openllama-3b/lora.yml
@@ -12,7 +12,7 @@ datasets:
 dataset_prepared_path:
 val_set_size: 0.02
 adapter: lora
-lora_model_dir:
+peft_model_dir:
 sequence_len: 1024
 sample_packing: true
 lora_r: 8
--- a/examples/openllama-3b/qlora.yml
+++ b/examples/openllama-3b/qlora.yml
@@ -12,7 +12,7 @@ datasets:
 dataset_prepared_path:
 val_set_size: 0.01
 adapter: qlora
-lora_model_dir:
+peft_model_dir:
 sequence_len: 1024
 sample_packing: true
 lora_r: 8
--- a/examples/phi/phi-ft.yml
+++ b/examples/phi/phi-ft.yml
@@ -22,7 +22,7 @@ sample_packing: true
 pad_to_sequence_len:

 adapter:
-lora_model_dir:
+peft_model_dir:
 lora_r:
 lora_alpha:
 lora_dropout:
--- a/examples/phi/phi-qlora.yml
+++ b/examples/phi/phi-qlora.yml
@@ -22,7 +22,7 @@ sample_packing: false  # not CURRENTLY compatible with LoRAs
 pad_to_sequence_len:

 adapter: qlora
-lora_model_dir:
+peft_model_dir:
 lora_r: 64
 lora_alpha: 32
 lora_dropout: 0.05
--- a/examples/pythia-12b/config.yml
+++ b/examples/pythia-12b/config.yml
@@ -13,7 +13,7 @@ datasets:
 dataset_prepared_path:
 val_set_size: 0.05
 adapter:
-lora_model_dir:
+peft_model_dir:
 sequence_len: 2048
 max_packed_sequence_len: 2048
 lora_r: 64
--- a/examples/pythia/lora.yml
+++ b/examples/pythia/lora.yml
@@ -7,7 +7,7 @@ datasets:
 dataset_prepared_path:
 val_set_size: 0.05
 adapter: lora
-lora_model_dir:
+peft_model_dir:
 sequence_len: 512
 lora_r: 16
 lora_alpha: 32
--- a/examples/redpajama/config-3b.yml
+++ b/examples/redpajama/config-3b.yml
@@ -10,7 +10,7 @@ datasets:
 dataset_prepared_path:
 val_set_size: 0.02
 adapter:
-lora_model_dir:
+peft_model_dir:
 sequence_len: 2048
 max_packed_sequence_len:
 lora_r: 8
--- a/examples/replit-3b/config-lora.yml
+++ b/examples/replit-3b/config-lora.yml
@@ -8,7 +8,7 @@ datasets:
 dataset_prepared_path:
 val_set_size: 0.05
 adapter: lora
-lora_model_dir:
+peft_model_dir:
 sequence_len: 2048
 max_packed_sequence_len:
 lora_r: 8
--- a/examples/xgen-7b/xgen-7b-8k-qlora.yml
+++ b/examples/xgen-7b/xgen-7b-8k-qlora.yml
@@ -20,7 +20,7 @@ dataset_prepared_path:
 val_set_size: 0.01
 # enable QLoRA
 adapter: qlora
-lora_model_dir:
+peft_model_dir:
 sequence_len: 8192
 max_packed_sequence_len: