Training in progress, epoch 10, checkpoint

checkpoint-1875/README.md

@@ -0,0 +1,9 @@
+---
+library_name: peft
+---
+## Training procedure
+
+### Framework versions
+
+
+- PEFT 0.5.0

checkpoint-1875/adapter_config.json

@@ -0,0 +1,25 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "Qwen/Qwen1.5-4B",
+  "bias": "none",
+  "fan_in_fan_out": false,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 16,
+  "lora_dropout": 0.05,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 16,
+  "revision": null,
+  "target_modules": [
+    "q_proj",
+    "k_proj",
+    "v_proj",
+    "down_proj",
+    "gate_proj",
+    "up_proj"
+  ],
+  "task_type": "CAUSAL_LM"
+}

checkpoint-1875/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2cde15f987c6d54d41fbb85be064a6a8bd3683ae1e750768207ff142a4c5ebf8
+size 112130416

checkpoint-1875/added_tokens.json

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

checkpoint-1875/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d40323bbf5633b2dbc080742dc6c7564edec1459de0e538c51b80cec4fba64a0
+size 224537202

checkpoint-1875/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc10eeac5d1e6bb4c7a1b387908144a2e0847830e26728f93552c9fb4d2e1720
+size 15024

checkpoint-1875/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c593af85db8b9470d75256b6595c2a05b9b1b3cf654bfe652261efc77f8485a3
+size 15024

checkpoint-1875/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2ab2a578ef6e7bc4fc6cd5fcc0728bf7269a844ba9f41dc4418b2c46654379e0
+size 15024

checkpoint-1875/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f94aba647d23a8ddb07996daaa779821991c91b0188b5c88eb2a50901be9984a
+size 15024

checkpoint-1875/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9b4d47d61c2c2894ca219f51a63b8fcc62fcd8679785d1c36cdc116da736d46
+size 1064

checkpoint-1875/special_tokens_map.json

@@ -0,0 +1,14 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>"
+  ],
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": "<|endoftext|>"
+}

checkpoint-1875/tokenizer_config.json

@@ -0,0 +1,43 @@
+{
+  "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": "{% for message in messages %}{% if loop.first and messages[0]['role'] != 'system' %}{{ '<|im_start|>system\nYou are a helpful assistant<|im_end|>\n' }}{% endif %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|endoftext|>",
+  "errors": "replace",
+  "model_max_length": 32768,
+  "pad_token": "<|endoftext|>",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}

checkpoint-1875/trainer_state.json

@@ -0,0 +1,309 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 10.0,
+  "eval_steps": 500,
+  "global_step": 1875,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.5333333333333333,
+      "grad_norm": 0.5194531679153442,
+      "learning_rate": 0.0003,
+      "loss": 1.7259,
+      "step": 100
+    },
+    {
+      "epoch": 0.9973333333333333,
+      "eval_accuracy": 0.5753333333333334,
+      "eval_loss": 1.6105104684829712,
+      "eval_runtime": 5.418,
+      "eval_samples_per_second": 92.285,
+      "eval_steps_per_second": 11.628,
+      "step": 187
+    },
+    {
+      "epoch": 0.9973333333333333,
+      "eval_exact_match": 17.2,
+      "eval_f1": 26.890173160173163,
+      "step": 187
+    },
+    {
+      "epoch": 1.0666666666666667,
+      "grad_norm": 0.6734262704849243,
+      "learning_rate": 0.0003,
+      "loss": 1.6156,
+      "step": 200
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 0.7422465682029724,
+      "learning_rate": 0.0003,
+      "loss": 1.3358,
+      "step": 300
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5722564102564103,
+      "eval_loss": 1.6464648246765137,
+      "eval_runtime": 5.1856,
+      "eval_samples_per_second": 96.422,
+      "eval_steps_per_second": 12.149,
+      "step": 375
+    },
+    {
+      "epoch": 2.0,
+      "eval_exact_match": 19.0,
+      "eval_f1": 27.762380952380955,
+      "step": 375
+    },
+    {
+      "epoch": 2.1333333333333333,
+      "grad_norm": 0.828364908695221,
+      "learning_rate": 0.0003,
+      "loss": 1.2607,
+      "step": 400
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.9025559425354004,
+      "learning_rate": 0.0003,
+      "loss": 0.9854,
+      "step": 500
+    },
+    {
+      "epoch": 2.997333333333333,
+      "eval_accuracy": 0.5707692307692308,
+      "eval_loss": 1.7620763778686523,
+      "eval_runtime": 5.2589,
+      "eval_samples_per_second": 95.077,
+      "eval_steps_per_second": 11.98,
+      "step": 562
+    },
+    {
+      "epoch": 2.997333333333333,
+      "eval_exact_match": 18.2,
+      "eval_f1": 27.53142857142857,
+      "step": 562
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 0.8217537999153137,
+      "learning_rate": 0.0003,
+      "loss": 0.9099,
+      "step": 600
+    },
+    {
+      "epoch": 3.7333333333333334,
+      "grad_norm": 0.8315110206604004,
+      "learning_rate": 0.0003,
+      "loss": 0.7508,
+      "step": 700
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5663076923076923,
+      "eval_loss": 1.9380683898925781,
+      "eval_runtime": 5.3338,
+      "eval_samples_per_second": 93.742,
+      "eval_steps_per_second": 11.811,
+      "step": 750
+    },
+    {
+      "epoch": 4.0,
+      "eval_exact_match": 16.0,
+      "eval_f1": 24.611904761904768,
+      "step": 750
+    },
+    {
+      "epoch": 4.266666666666667,
+      "grad_norm": 0.6763857007026672,
+      "learning_rate": 0.0003,
+      "loss": 0.6912,
+      "step": 800
+    },
+    {
+      "epoch": 4.8,
+      "grad_norm": 0.8408166766166687,
+      "learning_rate": 0.0003,
+      "loss": 0.6396,
+      "step": 900
+    },
+    {
+      "epoch": 4.997333333333334,
+      "eval_accuracy": 0.5634358974358974,
+      "eval_loss": 1.9925932884216309,
+      "eval_runtime": 5.3717,
+      "eval_samples_per_second": 93.08,
+      "eval_steps_per_second": 11.728,
+      "step": 937
+    },
+    {
+      "epoch": 4.997333333333334,
+      "eval_exact_match": 16.0,
+      "eval_f1": 24.313809523809528,
+      "step": 937
+    },
+    {
+      "epoch": 5.333333333333333,
+      "grad_norm": 0.6721271276473999,
+      "learning_rate": 0.0003,
+      "loss": 0.5959,
+      "step": 1000
+    },
+    {
+      "epoch": 5.866666666666667,
+      "grad_norm": 0.9258321523666382,
+      "learning_rate": 0.0003,
+      "loss": 0.5833,
+      "step": 1100
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5611794871794872,
+      "eval_loss": 2.101524829864502,
+      "eval_runtime": 5.4208,
+      "eval_samples_per_second": 92.238,
+      "eval_steps_per_second": 11.622,
+      "step": 1125
+    },
+    {
+      "epoch": 6.0,
+      "eval_exact_match": 17.0,
+      "eval_f1": 25.444285714285726,
+      "step": 1125
+    },
+    {
+      "epoch": 6.4,
+      "grad_norm": 0.6029295921325684,
+      "learning_rate": 0.0003,
+      "loss": 0.545,
+      "step": 1200
+    },
+    {
+      "epoch": 6.933333333333334,
+      "grad_norm": 0.5719736814498901,
+      "learning_rate": 0.0003,
+      "loss": 0.5567,
+      "step": 1300
+    },
+    {
+      "epoch": 6.997333333333334,
+      "eval_accuracy": 0.5607179487179487,
+      "eval_loss": 2.1645195484161377,
+      "eval_runtime": 5.1974,
+      "eval_samples_per_second": 96.202,
+      "eval_steps_per_second": 12.121,
+      "step": 1312
+    },
+    {
+      "epoch": 6.997333333333334,
+      "eval_exact_match": 15.0,
+      "eval_f1": 23.618095238095247,
+      "step": 1312
+    },
+    {
+      "epoch": 7.466666666666667,
+      "grad_norm": 0.49924588203430176,
+      "learning_rate": 0.0003,
+      "loss": 0.5205,
+      "step": 1400
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 0.5360875129699707,
+      "learning_rate": 0.0003,
+      "loss": 0.5411,
+      "step": 1500
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5614358974358974,
+      "eval_loss": 2.2039756774902344,
+      "eval_runtime": 5.0936,
+      "eval_samples_per_second": 98.162,
+      "eval_steps_per_second": 12.368,
+      "step": 1500
+    },
+    {
+      "epoch": 8.0,
+      "eval_exact_match": 16.8,
+      "eval_f1": 25.132380952380966,
+      "step": 1500
+    },
+    {
+      "epoch": 8.533333333333333,
+      "grad_norm": 0.44560280442237854,
+      "learning_rate": 0.0003,
+      "loss": 0.5028,
+      "step": 1600
+    },
+    {
+      "epoch": 8.997333333333334,
+      "eval_accuracy": 0.5608205128205128,
+      "eval_loss": 2.236473321914673,
+      "eval_runtime": 5.2122,
+      "eval_samples_per_second": 95.93,
+      "eval_steps_per_second": 12.087,
+      "step": 1687
+    },
+    {
+      "epoch": 8.997333333333334,
+      "eval_exact_match": 16.0,
+      "eval_f1": 24.712380952380958,
+      "step": 1687
+    },
+    {
+      "epoch": 9.066666666666666,
+      "grad_norm": 0.42847225069999695,
+      "learning_rate": 0.0003,
+      "loss": 0.5246,
+      "step": 1700
+    },
+    {
+      "epoch": 9.6,
+      "grad_norm": 0.4186452627182007,
+      "learning_rate": 0.0003,
+      "loss": 0.5041,
+      "step": 1800
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.5604615384615385,
+      "eval_loss": 2.2862467765808105,
+      "eval_runtime": 5.8141,
+      "eval_samples_per_second": 85.998,
+      "eval_steps_per_second": 10.836,
+      "step": 1875
+    },
+    {
+      "epoch": 10.0,
+      "eval_exact_match": 14.4,
+      "eval_f1": 22.683333333333344,
+      "step": 1875
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 9350,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 50,
+  "save_steps": 500,
+  "stateful_callbacks": {
+    "TrainerControl": {
+      "args": {
+        "should_epoch_stop": false,
+        "should_evaluate": false,
+        "should_log": false,
+        "should_save": true,
+        "should_training_stop": false
+      },
+      "attributes": {}
+    }
+  },
+  "total_flos": 3.265462178886451e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-1875/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a48aaa718d03f9c3a1cf4e80dbff36330a05dd346d42a3b722f1ae2626aabe3c
+size 5304