初始化项目，由ModelHub XC社区提供模型

Model: burtenshaw/Qwen1.5-0.5B-dpo-mix-7k
Source: Original Platform

checkpoint-313/config.json

@@ -0,0 +1,28 @@
+{
+  "_name_or_path": "Qwen/Qwen1.5-0.5B",
+  "architectures": [
+    "Qwen2ForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151643,
+  "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": "4.39.2",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "vocab_size": 151936
+}

checkpoint-313/generation_config.json

@@ -0,0 +1,6 @@
+{
+  "bos_token_id": 151643,
+  "eos_token_id": 151643,
+  "max_new_tokens": 2048,
+  "transformers_version": "4.39.2"
+}

checkpoint-313/model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bbe626c1033f0b0b6486b1203488cd6db2994c96f78d31b57e17e8262e80f3d1
+size 1239173352

checkpoint-313/optimizer.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc73358597b82368e59eabbe96fa657b47ccab1cef03cb3e63f43a80751855a6
+size 3712103162

checkpoint-313/pytorch_model_fsdp.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88628f8605e07512111aa6e00bd0c7736ac7d5b8fdd3f8c61c9e1860ee33e49c
+size 1239275353

checkpoint-313/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:308f94f9a5c24e1bad5c393d56ae7af7782600f4e791d9c6ac35b22fff2105b6
+size 15024

checkpoint-313/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b056f3c23cb32dc77a2ec9e7651e0b64e4440e21f0fdf969b86bfc56a1cbdf06
+size 15024

checkpoint-313/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f3f8a05714bc528f4885a2816181652f2303b3e8150f89b56aaee6bec56aa520
+size 15024

checkpoint-313/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f755bd3c330281961e5c03af9d10ce8c1e1678619d384f6f1fd5fd7dce2ff50
+size 15024

checkpoint-313/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93a3a847b2da5cc99667fb5595d3d91d25d426f2a3b3f1ec256909eaaddfc80b
+size 1064

checkpoint-313/trainer_state.json

@@ -0,0 +1,71 @@
+{
+  "best_metric": 1.1438744068145752,
+  "best_model_checkpoint": "./checkpoints/dpo-mix-7k/Qwen1.5-0.5B-dpo-mix-7k-lambda1.0-ORPO-29-9-49/checkpoint-313",
+  "epoch": 1.0,
+  "eval_steps": 500,
+  "global_step": 313,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.16,
+      "grad_norm": 616.0,
+      "learning_rate": 2.5e-06,
+      "loss": 15.2717,
+      "step": 50
+    },
+    {
+      "epoch": 0.32,
+      "grad_norm": 5.59375,
+      "learning_rate": 5e-06,
+      "loss": 4.0109,
+      "step": 100
+    },
+    {
+      "epoch": 0.48,
+      "grad_norm": 5.65625,
+      "learning_rate": 4.889350784534168e-06,
+      "loss": 1.2041,
+      "step": 150
+    },
+    {
+      "epoch": 0.64,
+      "grad_norm": 4.90625,
+      "learning_rate": 4.5671977372432355e-06,
+      "loss": 1.1597,
+      "step": 200
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 43.5,
+      "learning_rate": 4.062057643681335e-06,
+      "loss": 1.1542,
+      "step": 250
+    },
+    {
+      "epoch": 0.96,
+      "grad_norm": 9.8125,
+      "learning_rate": 3.4186451878908393e-06,
+      "loss": 1.1222,
+      "step": 300
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 1.1438744068145752,
+      "eval_runtime": 34.1359,
+      "eval_samples_per_second": 16.551,
+      "eval_steps_per_second": 1.055,
+      "step": 313
+    }
+  ],
+  "logging_steps": 50,
+  "max_steps": 626,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 2,
+  "save_steps": 500,
+  "total_flos": 4744684177981440.0,
+  "train_batch_size": 4,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-313/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ba636fcd247d639f405e2e00cd9717702cadada97437b5285147fee89e3f28e2
+size 5048