stuff

examples/llama-3/fft-8b-tp.yml

@@ -0,0 +1,58 @@
+base_model: NousResearch/Meta-Llama-3.1-8B
+
+load_in_8bit: false
+load_in_4bit: false
+strict: false
+
+datasets:
+  - path: tatsu-lab/alpaca
+    type: alpaca
+dataset_prepared_path: last_run_prepared
+val_set_size: 0.05
+output_dir: ./outputs/out
+
+sequence_len: 8192
+sample_packing: true
+pad_to_sequence_len: true
+
+wandb_project:
+wandb_entity:
+wandb_watch:
+wandb_name:
+wandb_log_model:
+
+gradient_accumulation_steps: 8
+micro_batch_size: 1
+num_epochs: 1
+optimizer: paged_adamw_8bit
+lr_scheduler: cosine
+learning_rate: 2e-5
+
+train_on_inputs: false
+group_by_length: false
+bf16: auto
+fp16:
+tf32: false
+
+tensor_parallel: 'auto'
+
+gradient_checkpointing: true
+gradient_checkpointing_kwargs:
+  use_reentrant: false
+early_stopping_patience:
+resume_from_checkpoint:
+logging_steps: 1
+xformers_attention:
+flash_attention: true
+
+warmup_steps: 100
+evals_per_epoch: 2
+eval_table_size:
+saves_per_epoch: 1
+debug:
+deepspeed:
+weight_decay: 0.0
+fsdp:
+fsdp_config:
+special_tokens:
+  pad_token: <|end_of_text|>

examples/llama-3/lora-8b-tp.yml

@@ -0,0 +1,73 @@
+base_model: NousResearch/Meta-Llama-3.1-8B
+model_type: LlamaForCausalLM
+tokenizer_type: AutoTokenizer
+
+load_in_8bit: true
+load_in_4bit: false
+strict: false
+
+datasets:
+  - path: mhenrichsen/alpaca_2k_test
+    type: alpaca
+dataset_prepared_path:
+val_set_size: 0.05
+output_dir: ./outputs/lora-out
+
+sequence_len: 4096
+sample_packing: true
+eval_sample_packing: false
+pad_to_sequence_len: true
+
+adapter: lora
+lora_model_dir:
+lora_r: 32
+lora_alpha: 16
+lora_dropout: 0.05
+lora_target_linear: true
+lora_fan_in_fan_out:
+lora_modules_to_save:
+  - embed_tokens
+  - lm_head
+
+wandb_project:
+wandb_entity:
+wandb_watch:
+wandb_name:
+wandb_log_model:
+
+gradient_accumulation_steps: 4
+micro_batch_size: 2
+num_epochs: 4
+optimizer: adamw_bnb_8bit
+lr_scheduler: cosine
+learning_rate: 0.0002
+
+train_on_inputs: false
+group_by_length: false
+bf16: auto
+fp16:
+tf32: false
+
+tensor_parallel: 'auto'
+
+gradient_checkpointing: true
+early_stopping_patience:
+resume_from_checkpoint:
+local_rank:
+logging_steps: 1
+xformers_attention:
+flash_attention: true
+s2_attention:
+
+warmup_steps: 10
+evals_per_epoch: 4
+eval_table_size:
+eval_max_new_tokens: 128
+saves_per_epoch: 1
+debug:
+deepspeed:
+weight_decay: 0.0
+fsdp:
+fsdp_config:
+special_tokens:
+   pad_token: <|end_of_text|>

src/axolotl/core/trainer_builder.py

@@ -1319,6 +1319,10 @@ class TrainerBuilderBase(abc.ABC):
         if hasattr(model, "add_model_tags"):
             model.add_model_tags(["axolotl"])
 
+        if self.cfg.tensor_parallel == "auto" and self.model.supports_tp_plan:
+            os.environ["ACCELERATE_USE_TP"] = "true"
+            # self.model =
+
     @property
     def model_ref(self):
         return self._model_ref

src/axolotl/utils/config/models/input/v0_4_1/__init__.py

@@ -393,7 +393,7 @@ class ModelInputConfig(BaseModel):
         default=None, json_schema_extra={"description": "transformers processor class"}
     )
     trust_remote_code: Optional[bool] = None
-
+    tensor_parallel: Optional[Union[Literal["auto"], bool]] = "auto"
     model_kwargs: Optional[Dict[str, Any]] = None
 
     @field_validator("trust_remote_code")

src/axolotl/utils/models.py

@@ -1187,9 +1187,15 @@ class ModelLoader:
             gc.collect()
             torch.cuda.empty_cache()
 
+        self.post_loading_set_env()
+
         # TODO resume_from_checkpoint handling
         return self.model, lora_config
 
+    def post_loading_set_env(self):
+        if self.cfg.tensor_parallel == "auto" and self.model.supports_tp_plan:
+            os.environ["ACCELERATE_USE_TP"] = "true"
+
 
 def load_model(
     cfg: DictDefault,