Upload 17 files

README.md

@@ -1,3 +1,59 @@
----
-license: wtfpl
----
+---
+license: other
+library_name: peft
+tags:
+- llama-factory
+- lora
+- generated_from_trainer
+base_model: Qwen/Qwen1.5-0.5B-Chat
+model-index:
+- name: QwenTT-0.5B-INT8
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+# QwenTT-0.5B-INT8
+
+This model is a fine-tuned version of [Qwen/Qwen1.5-0.5B-Chat](https://huggingface.co/Qwen/Qwen1.5-0.5B-Chat) on the identity dataset.
+
+## Model description
+
+More information needed
+
+## Intended uses & limitations
+
+More information needed
+
+## Training and evaluation data
+
+More information needed
+
+## Training procedure
+
+### Training hyperparameters
+
+The following hyperparameters were used during training:
+- learning_rate: 5e-05
+- train_batch_size: 2
+- eval_batch_size: 8
+- seed: 42
+- gradient_accumulation_steps: 8
+- total_train_batch_size: 16
+- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
+- lr_scheduler_type: cosine
+- num_epochs: 3.0
+- mixed_precision_training: Native AMP
+
+### Training results
+
+
+
+### Framework versions
+
+- PEFT 0.11.1
+- Transformers 4.40.2
+- Pytorch 2.2.1+cu121
+- Datasets 2.19.1
+- Tokenizers 0.19.1

added_tokens.json

@@ -0,0 +1,5 @@
+{
+  "<|endoftext|>": 151643,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644
+}

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.608695652173913,
+    "total_flos": 27871774801920.0,
+    "train_loss": 3.3604891459147135,
+    "train_runtime": 81.3934,
+    "train_samples_per_second": 3.354,
+    "train_steps_per_second": 0.184
+}

config.json

@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen1.5-0.5B-Chat",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layer_replication": null,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "loftq_config": {},
+  "lora_alpha": 16,
+  "lora_dropout": 0,
+  "megatron_config": null,
+  "megatron_core": "megatron.core",
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 8,
+  "rank_pattern": {},
+  "revision": null,
+  "target_modules": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bbe5be185b3c4eac11027a7757c117526bec685ef0b8f1b04cd85732532d1060
+size 3158328

running_log.txt

@@ -0,0 +1,165 @@
+05/19/2024 22:55:50 - INFO - transformers.tokenization_utils_base - loading file vocab.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen1.5-0.5B-Chat/snapshots/4d14e384a4b037942bb3f3016665157c8bcb70ea/vocab.json
+
+05/19/2024 22:55:50 - INFO - transformers.tokenization_utils_base - loading file merges.txt from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen1.5-0.5B-Chat/snapshots/4d14e384a4b037942bb3f3016665157c8bcb70ea/merges.txt
+
+05/19/2024 22:55:50 - INFO - transformers.tokenization_utils_base - loading file tokenizer.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen1.5-0.5B-Chat/snapshots/4d14e384a4b037942bb3f3016665157c8bcb70ea/tokenizer.json
+
+05/19/2024 22:55:50 - INFO - transformers.tokenization_utils_base - loading file added_tokens.json from cache at None
+
+05/19/2024 22:55:50 - INFO - transformers.tokenization_utils_base - loading file special_tokens_map.json from cache at None
+
+05/19/2024 22:55:50 - INFO - transformers.tokenization_utils_base - loading file tokenizer_config.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen1.5-0.5B-Chat/snapshots/4d14e384a4b037942bb3f3016665157c8bcb70ea/tokenizer_config.json
+
+05/19/2024 22:55:51 - WARNING - transformers.tokenization_utils_base - Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
+
+05/19/2024 22:55:51 - INFO - llamafactory.data.template - Replace eos token: <|im_end|>
+
+05/19/2024 22:55:51 - INFO - llamafactory.data.loader - Loading dataset identity.json...
+
+05/19/2024 22:55:58 - INFO - transformers.configuration_utils - loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen1.5-0.5B-Chat/snapshots/4d14e384a4b037942bb3f3016665157c8bcb70ea/config.json
+
+05/19/2024 22:55:58 - INFO - transformers.configuration_utils - Model config Qwen2Config {
+  "_name_or_path": "Qwen/Qwen1.5-0.5B-Chat",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 2816,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 21,
+  "model_type": "qwen2",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "num_key_value_heads": 16,
+  "rms_norm_eps": 1e-06,
+  "rope_theta": 1000000.0,
+  "sliding_window": 32768,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.40.2",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}
+
+
+05/19/2024 22:55:58 - INFO - llamafactory.model.utils.quantization - Quantizing model to 8 bit.
+
+05/19/2024 22:55:58 - INFO - transformers.modeling_utils - loading weights file model.safetensors from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen1.5-0.5B-Chat/snapshots/4d14e384a4b037942bb3f3016665157c8bcb70ea/model.safetensors
+
+05/19/2024 22:55:58 - INFO - transformers.modeling_utils - Instantiating Qwen2ForCausalLM model under default dtype torch.float16.
+
+05/19/2024 22:55:58 - INFO - transformers.generation.configuration_utils - Generate config GenerationConfig {
+  "bos_token_id": 151643,
+  "eos_token_id": 151645
+}
+
+
+05/19/2024 22:56:02 - INFO - transformers.modeling_utils - All model checkpoint weights were used when initializing Qwen2ForCausalLM.
+
+
+05/19/2024 22:56:02 - INFO - transformers.modeling_utils - All the weights of Qwen2ForCausalLM were initialized from the model checkpoint at Qwen/Qwen1.5-0.5B-Chat.
+If your task is similar to the task the model of the checkpoint was trained on, you can already use Qwen2ForCausalLM for predictions without further training.
+
+05/19/2024 22:56:02 - INFO - transformers.generation.configuration_utils - loading configuration file generation_config.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen1.5-0.5B-Chat/snapshots/4d14e384a4b037942bb3f3016665157c8bcb70ea/generation_config.json
+
+05/19/2024 22:56:02 - INFO - transformers.generation.configuration_utils - Generate config GenerationConfig {
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.1,
+  "top_p": 0.8
+}
+
+
+05/19/2024 22:56:02 - INFO - llamafactory.model.utils.checkpointing - Gradient checkpointing enabled.
+
+05/19/2024 22:56:02 - INFO - llamafactory.model.utils.attention - Using torch SDPA for faster training and inference.
+
+05/19/2024 22:56:02 - INFO - llamafactory.model.adapter - Upcasting trainable params to float32.
+
+05/19/2024 22:56:02 - INFO - llamafactory.model.adapter - Fine-tuning method: LoRA
+
+05/19/2024 22:56:02 - INFO - llamafactory.model.loader - trainable params: 786432 || all params: 464774144 || trainable%: 0.1692
+
+05/19/2024 22:56:02 - INFO - transformers.trainer - Using auto half precision backend
+
+05/19/2024 22:56:03 - INFO - transformers.trainer - ***** Running training *****
+
+05/19/2024 22:56:03 - INFO - transformers.trainer -   Num examples = 91
+
+05/19/2024 22:56:03 - INFO - transformers.trainer -   Num Epochs = 3
+
+05/19/2024 22:56:03 - INFO - transformers.trainer -   Instantaneous batch size per device = 2
+
+05/19/2024 22:56:03 - INFO - transformers.trainer -   Total train batch size (w. parallel, distributed & accumulation) = 16
+
+05/19/2024 22:56:03 - INFO - transformers.trainer -   Gradient Accumulation steps = 8
+
+05/19/2024 22:56:03 - INFO - transformers.trainer -   Total optimization steps = 15
+
+05/19/2024 22:56:03 - INFO - transformers.trainer -   Number of trainable parameters = 786,432
+
+05/19/2024 22:56:30 - INFO - llamafactory.extras.callbacks - {'loss': 3.4258, 'learning_rate': 3.7500e-05, 'epoch': 0.87}
+
+05/19/2024 22:56:57 - INFO - llamafactory.extras.callbacks - {'loss': 3.3578, 'learning_rate': 1.2500e-05, 'epoch': 1.74}
+
+05/19/2024 22:57:24 - INFO - llamafactory.extras.callbacks - {'loss': 3.2979, 'learning_rate': 0.0000e+00, 'epoch': 2.61}
+
+05/19/2024 22:57:24 - INFO - transformers.trainer - 
+
+Training completed. Do not forget to share your model on huggingface.co/models =)
+
+
+
+05/19/2024 22:57:24 - INFO - transformers.trainer - Saving model checkpoint to saves/Qwen1.5-0.5B-Chat/lora/QwenTT-0.5B-INT8
+
+05/19/2024 22:57:25 - INFO - transformers.configuration_utils - loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--Qwen--Qwen1.5-0.5B-Chat/snapshots/4d14e384a4b037942bb3f3016665157c8bcb70ea/config.json
+
+05/19/2024 22:57:25 - INFO - transformers.configuration_utils - Model config Qwen2Config {
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 1024,
+  "initializer_range": 0.02,
+  "intermediate_size": 2816,
+  "max_position_embeddings": 32768,
+  "max_window_layers": 21,
+  "model_type": "qwen2",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 24,
+  "num_key_value_heads": 16,
+  "rms_norm_eps": 1e-06,
+  "rope_theta": 1000000.0,
+  "sliding_window": 32768,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.40.2",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}
+
+
+05/19/2024 22:57:25 - INFO - transformers.tokenization_utils_base - tokenizer config file saved in saves/Qwen1.5-0.5B-Chat/lora/QwenTT-0.5B-INT8/tokenizer_config.json
+
+05/19/2024 22:57:25 - INFO - transformers.tokenization_utils_base - Special tokens file saved in saves/Qwen1.5-0.5B-Chat/lora/QwenTT-0.5B-INT8/special_tokens_map.json
+
+05/19/2024 22:57:25 - WARNING - llamafactory.extras.ploting - No metric eval_loss to plot.
+
+05/19/2024 22:57:25 - INFO - transformers.modelcard - Dropping the following result as it does not have all the necessary fields:
+{'task': {'name': 'Causal Language Modeling', 'type': 'text-generation'}}
+

special_tokens_map.json

@@ -0,0 +1,20 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,44 @@
+{
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{% set system_message = 'You are a helpful assistant.' %}{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{% if system_message is defined %}{{ '<|im_start|>system\n' + system_message + '<|im_end|>\n' }}{% endif %}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ '<|im_start|>user\n' + content + '<|im_end|>\n<|im_start|>assistant\n' }}{% elif message['role'] == 'assistant' %}{{ content + '<|im_end|>' + '\n' }}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "right",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.608695652173913,
+    "total_flos": 27871774801920.0,
+    "train_loss": 3.3604891459147135,
+    "train_runtime": 81.3934,
+    "train_samples_per_second": 3.354,
+    "train_steps_per_second": 0.184
+}

trainer_config.yaml

@@ -0,0 +1,31 @@
+cutoff_len: 1024
+dataset: identity
+dataset_dir: data
+do_train: true
+finetuning_type: lora
+flash_attn: auto
+fp16: true
+gradient_accumulation_steps: 8
+learning_rate: 5.0e-05
+logging_steps: 5
+lora_alpha: 16
+lora_dropout: 0
+lora_rank: 8
+lora_target: q_proj,v_proj
+lr_scheduler_type: cosine
+max_grad_norm: 1.0
+max_samples: 100000
+model_name_or_path: Qwen/Qwen1.5-0.5B-Chat
+num_train_epochs: 3.0
+optim: adamw_torch
+output_dir: saves/Qwen1.5-0.5B-Chat/lora/QwenTT-0.5B-INT8
+packing: false
+per_device_train_batch_size: 2
+plot_loss: true
+preprocessing_num_workers: 16
+quantization_bit: 8
+report_to: none
+save_steps: 100
+stage: sft
+template: qwen
+warmup_steps: 0

trainer_log.jsonl

@@ -0,0 +1,4 @@
+{"current_steps": 5, "total_steps": 15, "loss": 3.4258, "learning_rate": 3.7500000000000003e-05, "epoch": 0.8695652173913043, "percentage": 33.33, "elapsed_time": "0:00:27", "remaining_time": "0:00:54"}
+{"current_steps": 10, "total_steps": 15, "loss": 3.3578, "learning_rate": 1.2500000000000006e-05, "epoch": 1.7391304347826086, "percentage": 66.67, "elapsed_time": "0:00:54", "remaining_time": "0:00:27"}
+{"current_steps": 15, "total_steps": 15, "loss": 3.2979, "learning_rate": 0.0, "epoch": 2.608695652173913, "percentage": 100.0, "elapsed_time": "0:01:21", "remaining_time": "0:00:00"}
+{"current_steps": 15, "total_steps": 15, "epoch": 2.608695652173913, "percentage": 100.0, "elapsed_time": "0:01:21", "remaining_time": "0:00:00"}

trainer_state.json

@@ -0,0 +1,51 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 2.608695652173913,
+  "eval_steps": 500,
+  "global_step": 15,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.8695652173913043,
+      "grad_norm": 1.6034659147262573,
+      "learning_rate": 3.7500000000000003e-05,
+      "loss": 3.4258,
+      "step": 5
+    },
+    {
+      "epoch": 1.7391304347826086,
+      "grad_norm": 1.0319217443466187,
+      "learning_rate": 1.2500000000000006e-05,
+      "loss": 3.3578,
+      "step": 10
+    },
+    {
+      "epoch": 2.608695652173913,
+      "grad_norm": 1.2878139019012451,
+      "learning_rate": 0.0,
+      "loss": 3.2979,
+      "step": 15
+    },
+    {
+      "epoch": 2.608695652173913,
+      "step": 15,
+      "total_flos": 27871774801920.0,
+      "train_loss": 3.3604891459147135,
+      "train_runtime": 81.3934,
+      "train_samples_per_second": 3.354,
+      "train_steps_per_second": 0.184
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 15,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 3,
+  "save_steps": 100,
+  "total_flos": 27871774801920.0,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:44c4d8f46ae72d6f97f06fdd5f7f5b232339172b58af2232de6515d3688c44bf
+size 5176