Upload folder using huggingface_hub

README.md

@@ -0,0 +1,60 @@
+---
+license: other
+library_name: peft
+tags:
+- llama-factory
+- lora
+- unsloth
+- generated_from_trainer
+base_model: chujiezheng/Starling-LM-7B-alpha-ExPO
+model-index:
+- name: train_2024-05-08-19-49-29
+  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. -->
+
+# train_2024-05-08-19-49-29
+
+This model is a fine-tuned version of [chujiezheng/Starling-LM-7B-alpha-ExPO](https://huggingface.co/chujiezheng/Starling-LM-7B-alpha-ExPO) on the no_robots 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: 0.1
+- mixed_precision_training: Native AMP
+
+### Training results
+
+
+
+### Framework versions
+
+- PEFT 0.10.0
+- Transformers 4.40.1
+- Pytorch 2.2.1+cu121
+- Datasets 2.19.1
+- Tokenizers 0.19.1

adapter_config.json

@@ -0,0 +1,29 @@
+{
+  "alpha_pattern": {},
+  "auto_mapping": null,
+  "base_model_name_or_path": "chujiezheng/Starling-LM-7B-alpha-ExPO",
+  "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": "unsloth",
+  "target_modules": [
+    "v_proj",
+    "q_proj"
+  ],
+  "task_type": "CAUSAL_LM",
+  "use_dora": false,
+  "use_rslora": false
+}

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8e868c6f310cde5819d461685833d5c4ba914a779ab400a10df62f05a7ba90bc
+size 13648432

added_tokens.json

@@ -0,0 +1,5 @@
+{
+  "<sep>": 32002,
+  "<|end_of_turn|>": 32000,
+  "<|pad_0|>": 32001
+}

all_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.10105263157894737,
+    "total_flos": 1.783322464223232e+16,
+    "train_loss": 1.624589498837789,
+    "train_runtime": 968.353,
+    "train_samples_per_second": 0.981,
+    "train_steps_per_second": 0.062
+}

running_log.txt

@@ -0,0 +1,282 @@
+05/08/2024 19:50:37 - INFO - transformers.tokenization_utils_base - loading file tokenizer.model from cache at /root/.cache/huggingface/hub/models--chujiezheng--Starling-LM-7B-alpha-ExPO/snapshots/fc53fe6883ff3ea00a89be453a6cc586b3c369c3/tokenizer.model
+
+05/08/2024 19:50:37 - INFO - transformers.tokenization_utils_base - loading file tokenizer.json from cache at /root/.cache/huggingface/hub/models--chujiezheng--Starling-LM-7B-alpha-ExPO/snapshots/fc53fe6883ff3ea00a89be453a6cc586b3c369c3/tokenizer.json
+
+05/08/2024 19:50:37 - INFO - transformers.tokenization_utils_base - loading file added_tokens.json from cache at /root/.cache/huggingface/hub/models--chujiezheng--Starling-LM-7B-alpha-ExPO/snapshots/fc53fe6883ff3ea00a89be453a6cc586b3c369c3/added_tokens.json
+
+05/08/2024 19:50:37 - INFO - transformers.tokenization_utils_base - loading file special_tokens_map.json from cache at /root/.cache/huggingface/hub/models--chujiezheng--Starling-LM-7B-alpha-ExPO/snapshots/fc53fe6883ff3ea00a89be453a6cc586b3c369c3/special_tokens_map.json
+
+05/08/2024 19:50:37 - INFO - transformers.tokenization_utils_base - loading file tokenizer_config.json from cache at /root/.cache/huggingface/hub/models--chujiezheng--Starling-LM-7B-alpha-ExPO/snapshots/fc53fe6883ff3ea00a89be453a6cc586b3c369c3/tokenizer_config.json
+
+05/08/2024 19:50:38 - WARNING - transformers.tokenization_utils_base - Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
+
+05/08/2024 19:50:38 - INFO - llmtuner.data.loader - Loading dataset realign_no_robots.jsonl...
+
+05/08/2024 19:50:39 - INFO - transformers.configuration_utils - loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--chujiezheng--Starling-LM-7B-alpha-ExPO/snapshots/fc53fe6883ff3ea00a89be453a6cc586b3c369c3/config.json
+
+05/08/2024 19:50:39 - INFO - transformers.configuration_utils - Model config MistralConfig {
+  "_name_or_path": "chujiezheng/Starling-LM-7B-alpha-ExPO",
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 32000,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 8192,
+  "model_type": "mistral",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 10000.0,
+  "sliding_window": 4096,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.40.1",
+  "use_cache": true,
+  "vocab_size": 32002
+}
+
+
+05/08/2024 19:50:39 - INFO - transformers.configuration_utils - loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--chujiezheng--Starling-LM-7B-alpha-ExPO/snapshots/fc53fe6883ff3ea00a89be453a6cc586b3c369c3/config.json
+
+05/08/2024 19:50:39 - INFO - transformers.configuration_utils - Model config MistralConfig {
+  "_name_or_path": "chujiezheng/Starling-LM-7B-alpha-ExPO",
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 32000,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 8192,
+  "model_type": "mistral",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 10000.0,
+  "sliding_window": 4096,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.40.1",
+  "use_cache": true,
+  "vocab_size": 32002
+}
+
+
+05/08/2024 19:50:39 - INFO - transformers.configuration_utils - loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--chujiezheng--Starling-LM-7B-alpha-ExPO/snapshots/fc53fe6883ff3ea00a89be453a6cc586b3c369c3/config.json
+
+05/08/2024 19:50:39 - INFO - transformers.configuration_utils - Model config MistralConfig {
+  "_name_or_path": "chujiezheng/Starling-LM-7B-alpha-ExPO",
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 32000,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 8192,
+  "model_type": "mistral",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 10000.0,
+  "sliding_window": 4096,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.40.1",
+  "use_cache": true,
+  "vocab_size": 32002
+}
+
+
+05/08/2024 19:50:40 - INFO - transformers.configuration_utils - loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--chujiezheng--Starling-LM-7B-alpha-ExPO/snapshots/fc53fe6883ff3ea00a89be453a6cc586b3c369c3/config.json
+
+05/08/2024 19:50:40 - INFO - transformers.configuration_utils - Model config MistralConfig {
+  "_name_or_path": "chujiezheng/Starling-LM-7B-alpha-ExPO",
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 32000,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 8192,
+  "model_type": "mistral",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 10000.0,
+  "sliding_window": 4096,
+  "tie_word_embeddings": false,
+  "torch_dtype": "float16",
+  "transformers_version": "4.40.1",
+  "use_cache": true,
+  "vocab_size": 32002
+}
+
+
+05/08/2024 19:50:40 - INFO - transformers.modeling_utils - loading weights file model.safetensors from cache at /root/.cache/huggingface/hub/models--chujiezheng--Starling-LM-7B-alpha-ExPO/snapshots/fc53fe6883ff3ea00a89be453a6cc586b3c369c3/model.safetensors.index.json
+
+05/08/2024 19:50:40 - INFO - transformers.modeling_utils - Instantiating MistralForCausalLM model under default dtype torch.float16.
+
+05/08/2024 19:50:40 - INFO - transformers.generation.configuration_utils - Generate config GenerationConfig {
+  "bos_token_id": 1,
+  "eos_token_id": 32000
+}
+
+
+05/08/2024 19:51:52 - INFO - transformers.modeling_utils - All model checkpoint weights were used when initializing MistralForCausalLM.
+
+
+05/08/2024 19:51:52 - INFO - transformers.modeling_utils - All the weights of MistralForCausalLM were initialized from the model checkpoint at chujiezheng/Starling-LM-7B-alpha-ExPO.
+If your task is similar to the task the model of the checkpoint was trained on, you can already use MistralForCausalLM for predictions without further training.
+
+05/08/2024 19:51:52 - INFO - transformers.generation.configuration_utils - loading configuration file generation_config.json from cache at /root/.cache/huggingface/hub/models--chujiezheng--Starling-LM-7B-alpha-ExPO/snapshots/fc53fe6883ff3ea00a89be453a6cc586b3c369c3/generation_config.json
+
+05/08/2024 19:51:52 - INFO - transformers.generation.configuration_utils - Generate config GenerationConfig {
+  "bos_token_id": 1,
+  "do_sample": true,
+  "eos_token_id": 32000,
+  "max_length": 8192,
+  "pad_token_id": 0,
+  "temperature": 0.5
+}
+
+
+05/08/2024 19:51:52 - INFO - transformers.tokenization_utils_base - loading file tokenizer.model from cache at huggingface_tokenizers_cache/models--chujiezheng--Starling-LM-7B-alpha-ExPO/snapshots/fc53fe6883ff3ea00a89be453a6cc586b3c369c3/tokenizer.model
+
+05/08/2024 19:51:52 - INFO - transformers.tokenization_utils_base - loading file added_tokens.json from cache at huggingface_tokenizers_cache/models--chujiezheng--Starling-LM-7B-alpha-ExPO/snapshots/fc53fe6883ff3ea00a89be453a6cc586b3c369c3/added_tokens.json
+
+05/08/2024 19:51:52 - INFO - transformers.tokenization_utils_base - loading file special_tokens_map.json from cache at huggingface_tokenizers_cache/models--chujiezheng--Starling-LM-7B-alpha-ExPO/snapshots/fc53fe6883ff3ea00a89be453a6cc586b3c369c3/special_tokens_map.json
+
+05/08/2024 19:51:52 - INFO - transformers.tokenization_utils_base - loading file tokenizer_config.json from cache at huggingface_tokenizers_cache/models--chujiezheng--Starling-LM-7B-alpha-ExPO/snapshots/fc53fe6883ff3ea00a89be453a6cc586b3c369c3/tokenizer_config.json
+
+05/08/2024 19:51:52 - INFO - transformers.tokenization_utils_base - loading file tokenizer.json from cache at huggingface_tokenizers_cache/models--chujiezheng--Starling-LM-7B-alpha-ExPO/snapshots/fc53fe6883ff3ea00a89be453a6cc586b3c369c3/tokenizer.json
+
+05/08/2024 19:51:52 - WARNING - transformers.tokenization_utils_base - Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
+
+05/08/2024 19:51:53 - INFO - transformers.tokenization_utils_base - loading file tokenizer.model from cache at huggingface_tokenizers_cache/models--chujiezheng--Starling-LM-7B-alpha-ExPO/snapshots/fc53fe6883ff3ea00a89be453a6cc586b3c369c3/tokenizer.model
+
+05/08/2024 19:51:53 - INFO - transformers.tokenization_utils_base - loading file tokenizer.json from cache at huggingface_tokenizers_cache/models--chujiezheng--Starling-LM-7B-alpha-ExPO/snapshots/fc53fe6883ff3ea00a89be453a6cc586b3c369c3/tokenizer.json
+
+05/08/2024 19:51:53 - INFO - transformers.tokenization_utils_base - loading file added_tokens.json from cache at huggingface_tokenizers_cache/models--chujiezheng--Starling-LM-7B-alpha-ExPO/snapshots/fc53fe6883ff3ea00a89be453a6cc586b3c369c3/added_tokens.json
+
+05/08/2024 19:51:53 - INFO - transformers.tokenization_utils_base - loading file special_tokens_map.json from cache at huggingface_tokenizers_cache/models--chujiezheng--Starling-LM-7B-alpha-ExPO/snapshots/fc53fe6883ff3ea00a89be453a6cc586b3c369c3/special_tokens_map.json
+
+05/08/2024 19:51:53 - INFO - transformers.tokenization_utils_base - loading file tokenizer_config.json from cache at huggingface_tokenizers_cache/models--chujiezheng--Starling-LM-7B-alpha-ExPO/snapshots/fc53fe6883ff3ea00a89be453a6cc586b3c369c3/tokenizer_config.json
+
+05/08/2024 19:51:53 - WARNING - transformers.tokenization_utils_base - Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained.
+
+05/08/2024 19:51:53 - WARNING - transformers.models.llama.modeling_llama - chujiezheng/Starling-LM-7B-alpha-ExPO does not have a padding token! Will use pad_token = <unk>.
+
+05/08/2024 19:51:54 - INFO - llmtuner.model.utils.checkpointing - Gradient checkpointing enabled.
+
+05/08/2024 19:51:54 - INFO - llmtuner.model.adapter - Fine-tuning method: LoRA
+
+05/08/2024 19:51:54 - WARNING - transformers.models.llama.modeling_llama - Unsloth cannot patch MLP layers with our manual autograd engine since either LoRA adapters
+are not enabled or a bias term (like in Qwen) is used.
+
+05/08/2024 19:51:54 - WARNING - transformers.models.llama.modeling_llama - Unsloth cannot patch Attention layers with our manual autograd engine since either LoRA adapters
+are not enabled or a bias term (like in Qwen) is used.
+
+05/08/2024 19:51:54 - WARNING - transformers.models.llama.modeling_llama - Unsloth cannot patch O projection layer with our manual autograd engine since either LoRA adapters
+are not enabled or a bias term (like in Qwen) is used.
+
+05/08/2024 19:51:54 - WARNING - transformers.models.llama.modeling_llama - Unsloth 2024.4 patched 32 layers with 0 QKV layers, 0 O layers and 0 MLP layers.
+
+05/08/2024 19:51:54 - INFO - llmtuner.model.loader - trainable params: 3407872 || all params: 7245156352 || trainable%: 0.0470
+
+05/08/2024 19:51:54 - INFO - transformers.trainer - Using auto half precision backend
+
+05/08/2024 19:51:54 - WARNING - transformers.trainer - ==((====))==  Unsloth - 2x faster free finetuning | Num GPUs = 1
+   \\   /|    Num examples = 9,500 | Num Epochs = 1
+O^O/ \_/ \    Batch size per device = 2 | Gradient Accumulation steps = 8
+\        /    Total batch size = 16 | Total steps = 60
+ "-____-"     Number of trainable parameters = 3,407,872
+
+05/08/2024 19:53:17 - INFO - llmtuner.extras.callbacks - {'loss': 1.8162, 'learning_rate': 4.9148e-05, 'epoch': 0.01}
+
+05/08/2024 19:54:37 - INFO - llmtuner.extras.callbacks - {'loss': 1.7065, 'learning_rate': 4.6651e-05, 'epoch': 0.02}
+
+05/08/2024 19:55:55 - INFO - llmtuner.extras.callbacks - {'loss': 1.6809, 'learning_rate': 4.2678e-05, 'epoch': 0.03}
+
+05/08/2024 19:57:22 - INFO - llmtuner.extras.callbacks - {'loss': 1.7278, 'learning_rate': 3.7500e-05, 'epoch': 0.03}
+
+05/08/2024 19:58:49 - INFO - llmtuner.extras.callbacks - {'loss': 1.5573, 'learning_rate': 3.1470e-05, 'epoch': 0.04}
+
+05/08/2024 20:00:11 - INFO - llmtuner.extras.callbacks - {'loss': 1.5858, 'learning_rate': 2.5000e-05, 'epoch': 0.05}
+
+05/08/2024 20:01:39 - INFO - llmtuner.extras.callbacks - {'loss': 1.4219, 'learning_rate': 1.8530e-05, 'epoch': 0.06}
+
+05/08/2024 20:02:51 - INFO - llmtuner.extras.callbacks - {'loss': 1.6545, 'learning_rate': 1.2500e-05, 'epoch': 0.07}
+
+05/08/2024 20:04:11 - INFO - llmtuner.extras.callbacks - {'loss': 1.5896, 'learning_rate': 7.3223e-06, 'epoch': 0.08}
+
+05/08/2024 20:05:24 - INFO - llmtuner.extras.callbacks - {'loss': 1.4727, 'learning_rate': 3.3494e-06, 'epoch': 0.08}
+
+05/08/2024 20:06:42 - INFO - llmtuner.extras.callbacks - {'loss': 1.6785, 'learning_rate': 8.5185e-07, 'epoch': 0.09}
+
+05/08/2024 20:08:03 - INFO - llmtuner.extras.callbacks - {'loss': 1.6034, 'learning_rate': 0.0000e+00, 'epoch': 0.10}
+
+05/08/2024 20:08:03 - INFO - transformers.trainer - 
+
+Training completed. Do not forget to share your model on huggingface.co/models =)
+
+
+
+05/08/2024 20:08:03 - INFO - transformers.trainer - Saving model checkpoint to saves/OpenChat3.5-7B-Chat/lora/train_2024-05-08-19-49-29
+
+05/08/2024 20:08:03 - INFO - transformers.configuration_utils - loading configuration file config.json from cache at /root/.cache/huggingface/hub/models--chujiezheng--Starling-LM-7B-alpha-ExPO/snapshots/fc53fe6883ff3ea00a89be453a6cc586b3c369c3/config.json
+
+05/08/2024 20:08:03 - INFO - transformers.configuration_utils - Model config MistralConfig {
+  "_name_or_path": "openchat/openchat_3.5",
+  "architectures": [
+    "MistralForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 1,
+  "eos_token_id": 32000,
+  "hidden_act": "silu",
+  "hidden_size": 4096,
+  "initializer_range": 0.02,
+  "intermediate_size": 14336,
+  "max_position_embeddings": 8192,
+  "model_type": "mistral",
+  "num_attention_heads": 32,
+  "num_hidden_layers": 32,
+  "num_key_value_heads": 8,
+  "rms_norm_eps": 1e-05,
+  "rope_theta": 10000.0,
+  "sliding_window": 4096,
+  "tie_word_embeddings": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.40.1",
+  "use_cache": true,
+  "vocab_size": 32002
+}
+
+
+05/08/2024 20:08:03 - INFO - transformers.tokenization_utils_base - tokenizer config file saved in saves/OpenChat3.5-7B-Chat/lora/train_2024-05-08-19-49-29/tokenizer_config.json
+
+05/08/2024 20:08:03 - INFO - transformers.tokenization_utils_base - Special tokens file saved in saves/OpenChat3.5-7B-Chat/lora/train_2024-05-08-19-49-29/special_tokens_map.json
+
+05/08/2024 20:08:03 - 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 +1,41 @@
+{
+  "additional_special_tokens": [
+    "<|end_of_turn|>",
+    "<|pad_0|>"
+  ],
+  "bos_token": {
+    "content": "<s>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|end_of_turn|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|end_of_turn|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "sep_token": {
+    "content": "<sep>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "unk_token": {
+    "content": "<unk>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.model

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dadfd56d766715c61d2ef780a525ab43b8e6da4de6865bda3d95fdef5e134055
+size 493443

tokenizer_config.json

@@ -0,0 +1,73 @@
+{
+  "add_bos_token": true,
+  "add_eos_token": false,
+  "added_tokens_decoder": {
+    "0": {
+      "content": "<unk>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "1": {
+      "content": "<s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "2": {
+      "content": "</s>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32000": {
+      "content": "<|end_of_turn|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32001": {
+      "content": "<|pad_0|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "32002": {
+      "content": "<sep>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "additional_special_tokens": [
+    "<|end_of_turn|>",
+    "<|pad_0|>"
+  ],
+  "bos_token": "<s>",
+  "chat_template": "{% if messages[0]['role'] == 'system' %}{% set system_message = messages[0]['content'] %}{% endif %}{{ '<s>' + system_message }}{% for message in messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ 'GPT4 Correct User: ' + content + '<|end_of_turn|>' + 'GPT4 Correct Assistant:' }}{% elif message['role'] == 'assistant' %}{{ content + '<|end_of_turn|>' }}{% endif %}{% endfor %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|end_of_turn|>",
+  "legacy": true,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|end_of_turn|>",
+  "padding_side": "right",
+  "sep_token": "<sep>",
+  "sp_model_kwargs": {},
+  "spaces_between_special_tokens": false,
+  "split_special_tokens": false,
+  "tokenizer_class": "LlamaTokenizer",
+  "unk_token": "<unk>",
+  "use_default_system_prompt": true
+}

train_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.10105263157894737,
+    "total_flos": 1.783322464223232e+16,
+    "train_loss": 1.624589498837789,
+    "train_runtime": 968.353,
+    "train_samples_per_second": 0.981,
+    "train_steps_per_second": 0.062
+}

trainer_config.yaml

@@ -0,0 +1,29 @@
+cutoff_len: 1024
+dataset: no_robots
+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: chujiezheng/Starling-LM-7B-alpha-ExPO
+num_train_epochs: 0.1
+optim: adamw_torch
+output_dir: saves/OpenChat3.5-7B-Chat/lora/train_2024-05-08-19-49-29
+packing: false
+per_device_train_batch_size: 2
+report_to: none
+save_steps: 100
+stage: sft
+template: openchat
+use_unsloth: true
+warmup_steps: 0

trainer_log.jsonl

@@ -0,0 +1,13 @@
+{"current_steps": 5, "total_steps": 60, "loss": 1.8162, "learning_rate": 4.914814565722671e-05, "epoch": 0.008421052631578947, "percentage": 8.33, "elapsed_time": "0:01:22", "remaining_time": "0:15:10"}
+{"current_steps": 10, "total_steps": 60, "loss": 1.7065, "learning_rate": 4.665063509461097e-05, "epoch": 0.016842105263157894, "percentage": 16.67, "elapsed_time": "0:02:42", "remaining_time": "0:13:32"}
+{"current_steps": 15, "total_steps": 60, "loss": 1.6809, "learning_rate": 4.267766952966369e-05, "epoch": 0.02526315789473684, "percentage": 25.0, "elapsed_time": "0:04:00", "remaining_time": "0:12:01"}
+{"current_steps": 20, "total_steps": 60, "loss": 1.7278, "learning_rate": 3.7500000000000003e-05, "epoch": 0.03368421052631579, "percentage": 33.33, "elapsed_time": "0:05:27", "remaining_time": "0:10:54"}
+{"current_steps": 25, "total_steps": 60, "loss": 1.5573, "learning_rate": 3.147047612756302e-05, "epoch": 0.042105263157894736, "percentage": 41.67, "elapsed_time": "0:06:54", "remaining_time": "0:09:40"}
+{"current_steps": 30, "total_steps": 60, "loss": 1.5858, "learning_rate": 2.5e-05, "epoch": 0.05052631578947368, "percentage": 50.0, "elapsed_time": "0:08:16", "remaining_time": "0:08:16"}
+{"current_steps": 35, "total_steps": 60, "loss": 1.4219, "learning_rate": 1.852952387243698e-05, "epoch": 0.05894736842105263, "percentage": 58.33, "elapsed_time": "0:09:44", "remaining_time": "0:06:57"}
+{"current_steps": 40, "total_steps": 60, "loss": 1.6545, "learning_rate": 1.2500000000000006e-05, "epoch": 0.06736842105263158, "percentage": 66.67, "elapsed_time": "0:10:57", "remaining_time": "0:05:28"}
+{"current_steps": 45, "total_steps": 60, "loss": 1.5896, "learning_rate": 7.3223304703363135e-06, "epoch": 0.07578947368421053, "percentage": 75.0, "elapsed_time": "0:12:16", "remaining_time": "0:04:05"}
+{"current_steps": 50, "total_steps": 60, "loss": 1.4727, "learning_rate": 3.3493649053890326e-06, "epoch": 0.08421052631578947, "percentage": 83.33, "elapsed_time": "0:13:29", "remaining_time": "0:02:41"}
+{"current_steps": 55, "total_steps": 60, "loss": 1.6785, "learning_rate": 8.51854342773295e-07, "epoch": 0.09263157894736843, "percentage": 91.67, "elapsed_time": "0:14:47", "remaining_time": "0:01:20"}
+{"current_steps": 60, "total_steps": 60, "loss": 1.6034, "learning_rate": 0.0, "epoch": 0.10105263157894737, "percentage": 100.0, "elapsed_time": "0:16:08", "remaining_time": "0:00:00"}
+{"current_steps": 60, "total_steps": 60, "epoch": 0.10105263157894737, "percentage": 100.0, "elapsed_time": "0:16:08", "remaining_time": "0:00:00"}

trainer_state.json

@@ -0,0 +1,114 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 0.10105263157894737,
+  "eval_steps": 500,
+  "global_step": 60,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.008421052631578947,
+      "grad_norm": 1.9215331077575684,
+      "learning_rate": 4.914814565722671e-05,
+      "loss": 1.8162,
+      "step": 5
+    },
+    {
+      "epoch": 0.016842105263157894,
+      "grad_norm": 1.746415376663208,
+      "learning_rate": 4.665063509461097e-05,
+      "loss": 1.7065,
+      "step": 10
+    },
+    {
+      "epoch": 0.02526315789473684,
+      "grad_norm": 1.2568161487579346,
+      "learning_rate": 4.267766952966369e-05,
+      "loss": 1.6809,
+      "step": 15
+    },
+    {
+      "epoch": 0.03368421052631579,
+      "grad_norm": 1.087218165397644,
+      "learning_rate": 3.7500000000000003e-05,
+      "loss": 1.7278,
+      "step": 20
+    },
+    {
+      "epoch": 0.042105263157894736,
+      "grad_norm": 0.935480535030365,
+      "learning_rate": 3.147047612756302e-05,
+      "loss": 1.5573,
+      "step": 25
+    },
+    {
+      "epoch": 0.05052631578947368,
+      "grad_norm": 1.06438148021698,
+      "learning_rate": 2.5e-05,
+      "loss": 1.5858,
+      "step": 30
+    },
+    {
+      "epoch": 0.05894736842105263,
+      "grad_norm": 1.0331830978393555,
+      "learning_rate": 1.852952387243698e-05,
+      "loss": 1.4219,
+      "step": 35
+    },
+    {
+      "epoch": 0.06736842105263158,
+      "grad_norm": 0.7302573919296265,
+      "learning_rate": 1.2500000000000006e-05,
+      "loss": 1.6545,
+      "step": 40
+    },
+    {
+      "epoch": 0.07578947368421053,
+      "grad_norm": 0.8324851989746094,
+      "learning_rate": 7.3223304703363135e-06,
+      "loss": 1.5896,
+      "step": 45
+    },
+    {
+      "epoch": 0.08421052631578947,
+      "grad_norm": 1.0173665285110474,
+      "learning_rate": 3.3493649053890326e-06,
+      "loss": 1.4727,
+      "step": 50
+    },
+    {
+      "epoch": 0.09263157894736843,
+      "grad_norm": 0.763109564781189,
+      "learning_rate": 8.51854342773295e-07,
+      "loss": 1.6785,
+      "step": 55
+    },
+    {
+      "epoch": 0.10105263157894737,
+      "grad_norm": 0.8254455327987671,
+      "learning_rate": 0.0,
+      "loss": 1.6034,
+      "step": 60
+    },
+    {
+      "epoch": 0.10105263157894737,
+      "step": 60,
+      "total_flos": 1.783322464223232e+16,
+      "train_loss": 1.624589498837789,
+      "train_runtime": 968.353,
+      "train_samples_per_second": 0.981,
+      "train_steps_per_second": 0.062
+    }
+  ],
+  "logging_steps": 5,
+  "max_steps": 60,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 100,
+  "total_flos": 1.783322464223232e+16,
+  "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 sha256:8d7813b595d6dc62dc0132620651454ef332f506a8191f64fde55da56335e1af
+size 5176