axolotl example (#2784)

* axolotl example * Update llm/axolotl/axolotl-spot.yaml Co-authored-by: Zongheng Yang <[email protected]> * Update llm/axolotl/axolotl-spot.yaml Co-authored-by: Zongheng Yang <[email protected]> * Update llm/axolotl/axolotl-spot.yaml Co-authored-by: Zongheng Yang <[email protected]> * Apply suggestions from code review Co-authored-by: Zongheng Yang <[email protected]> * Apply suggestions from code review Co-authored-by: Zongheng Yang <[email protected]> * comment fixes --------- Co-authored-by: Zongheng Yang <[email protected]>
skypilot-org · Nov 15, 2023 · e6c690d · e6c690d
1 parent 80b2033
commit e6c690d
Show file tree

Hide file tree

Showing 5 changed files with 272 additions and 0 deletions.
diff --git a/llm/axolotl/axolotl-spot.yaml b/llm/axolotl/axolotl-spot.yaml
@@ -0,0 +1,48 @@
+# Usage:
+#
+#  Unmanaged spot (no auto-recovery; for debugging):
+#   HF_TOKEN=abc BUCKET=<unique-name> sky launch -c axolotl-spot axolotl-spot.yaml --env HF_TOKEN --env BUCKET -i30 --down
+#
+#  Managed spot (auto-recovery; for full runs):
+#   HF_TOKEN=abc BUCKET=<unique-name> sky spot launch -n axolotl-spot axolotl-spot.yaml --env HF_TOKEN --env BUCKET
+
+name: axolotl
+
+resources:
+  accelerators: A100:1
+  cloud: gcp # optional
+  use_spot: True
+
+workdir: mistral
+
+file_mounts:
+  /sky-notebook:
+    name: ${BUCKET}
+    mode: MOUNT
+
+setup: |
+  docker pull winglian/axolotl:main-py3.10-cu118-2.0.1
+
+run: |
+  docker run --gpus all \
+    -v ~/sky_workdir:/sky_workdir \
+    -v /root/.cache:/root/.cache \
+    winglian/axolotl:main-py3.10-cu118-2.0.1 \
+    huggingface-cli login --token ${HF_TOKEN}
+  
+  docker run --gpus all \
+    -v ~/sky_workdir:/sky_workdir \
+    -v /root/.cache:/root/.cache \
+    -v /sky-notebook:/sky-notebook \
+    winglian/axolotl:main-py3.10-cu118-2.0.1 \
+    accelerate launch -m axolotl.cli.train /sky_workdir/qlora-checkpoint.yaml
+
+envs:
+  HF_TOKEN: <your-huggingface-token> # TODO: Replace with huggingface token
+  BUCKET: <a-unique-bucket-name-to-use>
+
+
+
+
+
+
diff --git a/llm/axolotl/axolotl.yaml b/llm/axolotl/axolotl.yaml
@@ -0,0 +1,35 @@
+# Usage:
+#   HF_TOKEN=abc sky launch -c axolotl axolotl.yaml --env HF_TOKEN -y -i30 --down
+
+name: axolotl
+
+resources:
+  accelerators: L4:1
+  cloud: gcp # optional
+
+workdir: mistral
+
+setup: |
+  docker pull winglian/axolotl:main-py3.10-cu118-2.0.1
+
+run: |
+  docker run --gpus all \
+    -v ~/sky_workdir:/sky_workdir \
+    -v /root/.cache:/root/.cache \
+    winglian/axolotl:main-py3.10-cu118-2.0.1 \
+    huggingface-cli login --token ${HF_TOKEN} 
+
+  docker run --gpus all \
+    -v ~/sky_workdir:/sky_workdir \
+    -v /root/.cache:/root/.cache \
+    winglian/axolotl:main-py3.10-cu118-2.0.1 \
+    accelerate launch -m axolotl.cli.train /sky_workdir/qlora.yaml
+
+envs:
+  HF_TOKEN: <your-huggingface-token> # TODO: Replace with huggingface token
+
+
+
+
+
+
diff --git a/llm/axolotl/mistral/qlora-checkpoint.yaml b/llm/axolotl/mistral/qlora-checkpoint.yaml
@@ -0,0 +1,84 @@
+base_model: mistralai/Mistral-7B-v0.1
+model_type: MistralForCausalLM
+tokenizer_type: LlamaTokenizer
+is_mistral_derived_model: true
+
+load_in_8bit: false
+load_in_4bit: true
+strict: false
+
+datasets:
+  - path: mhenrichsen/alpaca_2k_test
+    type: alpaca
+
+dataset_prepared_path: /sky-notebook/alpaca_2k_test/last_run_prepared
+val_set_size: 0.05
+output_dir: /sky-notebook/alpaca_2k_test
+
+# hub_model_id: manishiitg/mistral-alpaca_2k_test # TODO: Replace with hub model id
+# hf_use_auth_token: false # TODO: push as private or public model
+
+adapter: qlora
+lora_model_dir:
+
+sequence_len: 8192
+sample_packing: true
+pad_to_sequence_len: true
+
+lora_r: 32
+lora_alpha: 16
+lora_dropout: 0.05
+lora_target_linear: true
+lora_fan_in_fan_out:
+lora_target_modules:
+  - gate_proj
+  - down_proj
+  - up_proj
+  - q_proj
+  - v_proj
+  - k_proj
+  - o_proj
+
+wandb_project:
+wandb_entity:
+wandb_watch:
+wandb_run_id:
+wandb_log_model:
+
+gradient_accumulation_steps: 4
+micro_batch_size: 2
+num_epochs: 1
+optimizer: adamw_bnb_8bit
+lr_scheduler: cosine
+learning_rate: 0.0002
+
+train_on_inputs: false
+group_by_length: false
+bf16: true
+fp16: false
+tf32: false
+
+gradient_checkpointing: true
+early_stopping_patience:
+resume_from_checkpoint:
+auto_resume_from_checkpoints: true ## manage check point resume from here
+local_rank:
+logging_steps: 1
+xformers_attention:
+flash_attention: true
+
+warmup_steps: 10
+eval_steps: 0.05
+eval_table_size:
+eval_table_max_new_tokens: 128
+save_steps: 2 ## increase based on your dataset
+save_strategy: steps
+debug:
+deepspeed:
+weight_decay: 0.0
+fsdp:
+fsdp_config:
+special_tokens:
+  bos_token: "<s>"
+  eos_token: "</s>"
+  unk_token: "<unk>"
diff --git a/llm/axolotl/mistral/qlora.yaml b/llm/axolotl/mistral/qlora.yaml
@@ -0,0 +1,81 @@
+base_model: mistralai/Mistral-7B-v0.1
+model_type: MistralForCausalLM
+tokenizer_type: LlamaTokenizer
+is_mistral_derived_model: true
+
+load_in_8bit: false
+load_in_4bit: true
+strict: false
+
+datasets:
+  - path: mhenrichsen/alpaca_2k_test
+    type: alpaca
+dataset_prepared_path: last_run_prepared
+val_set_size: 0.05
+output_dir: ./qlora-out
+
+# hub_model_id: manishiitg/mhenrichsen-alpaca_2k_test # TODO: Replace with hub model id
+# hf_use_auth_token: false # TODO: push as private or public model
+
+adapter: qlora
+lora_model_dir:
+
+sequence_len: 8192
+sample_packing: true
+pad_to_sequence_len: true
+
+lora_r: 32
+lora_alpha: 16
+lora_dropout: 0.05
+lora_target_linear: true
+lora_fan_in_fan_out:
+lora_target_modules:
+  - gate_proj
+  - down_proj
+  - up_proj
+  - q_proj
+  - v_proj
+  - k_proj
+  - o_proj
+
+wandb_project:
+wandb_entity:
+wandb_watch:
+wandb_run_id:
+wandb_log_model:
+
+gradient_accumulation_steps: 4
+micro_batch_size: 2
+num_epochs: 1
+optimizer: adamw_bnb_8bit
+lr_scheduler: cosine
+learning_rate: 0.0002
+
+train_on_inputs: false
+group_by_length: false
+bf16: true
+fp16: false
+tf32: false
+
+gradient_checkpointing: true
+early_stopping_patience:
+resume_from_checkpoint:
+local_rank:
+logging_steps: 1
+xformers_attention:
+flash_attention: true
+
+warmup_steps: 10
+eval_steps: 0.05
+eval_table_size:
+eval_table_max_new_tokens: 128
+save_steps:
+debug:
+deepspeed:
+weight_decay: 0.0
+fsdp:
+fsdp_config:
+special_tokens:
+  bos_token: "<s>"
+  eos_token: "</s>"
+  unk_token: "<unk>"
diff --git a/llm/axolotl/readme.md b/llm/axolotl/readme.md
@@ -0,0 +1,24 @@
+This example is using the same config located https://github.com/OpenAccess-AI-Collective/axolotl/tree/main/examples/mistral
+
+
+
+Simple training example
+```
+HF_TOKEN=abc sky launch -c axolotl axolotl.yaml --env HF_TOKEN -y -i30 --down
+ssh -L 8888:localhost:8888 axolotl
+sky down axolotl -y
+```
+
+
+
+To launch an unmanaged spot instance (no auto-recovery; good for debugging)
+```
+HF_TOKEN=abc BUCKET=<unique-name> sky launch -c axolotl-spot axolotl-spot.yaml --env HF_TOKEN --env BUCKET -i30 --down
+ssh -L 8888:localhost:8888 axolotl-spot
+```
+
+
+Launch managed spot instances (auto-recovery; for full runs):
+```
+HF_TOKEN=abc BUCKET=<unique-name> sky spot launch -n axolotl-spot axolotl-spot.yaml --env HF_TOKEN --env BUCKET
+```