Training in progress, epoch 6, checkpoint

checkpoint-1125/README.md

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

checkpoint-1125/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-1125/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c28bdd531b217293408d5f3107b66263e281276afd1059c84d0c6646330659a9
+size 112130416

checkpoint-1125/added_tokens.json

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

checkpoint-1125/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:92a2ca64062c6b7b119f4741e2956216eb2395b800b9f9f84f72898af477fac3
+size 224537202

checkpoint-1125/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b03aa0f99a630148793785cc4368b9aa39d285112b51d4618ef7f84233678643
+size 15024

checkpoint-1125/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c4a38fc0e609229f878ea72302c97e7a85fb7d5f2c0722355fc0ca831d7b90a
+size 15024

checkpoint-1125/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c5c333f9bbd832837f093bbeab47f839c90e56b5a6d62f9a8703ef99c9e7bb1b
+size 15024

checkpoint-1125/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a1f01ab80eacf848b47e599a17a9550f44f5e9d6aab27edd1b5198bfc3941f7c
+size 15024

checkpoint-1125/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:008acf78f82b9b6097eeb621787541b7684f43ba99c06c87e4aa184fe00fd9ee
+size 1064

checkpoint-1125/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-1125/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-1125/trainer_state.json

@@ -0,0 +1,200 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 6.0,
+  "eval_steps": 500,
+  "global_step": 1125,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.5333333333333333,
+      "grad_norm": 0.64143306016922,
+      "learning_rate": 0.0001,
+      "loss": 1.7779,
+      "step": 100
+    },
+    {
+      "epoch": 0.9973333333333333,
+      "eval_accuracy": 0.5734871794871795,
+      "eval_loss": 1.6191272735595703,
+      "eval_runtime": 5.393,
+      "eval_samples_per_second": 92.713,
+      "eval_steps_per_second": 11.682,
+      "step": 187
+    },
+    {
+      "epoch": 0.9973333333333333,
+      "eval_exact_match": 18.0,
+      "eval_f1": 28.280476190476193,
+      "step": 187
+    },
+    {
+      "epoch": 1.0666666666666667,
+      "grad_norm": 0.7118939161300659,
+      "learning_rate": 0.0001,
+      "loss": 1.6349,
+      "step": 200
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 1.0694799423217773,
+      "learning_rate": 0.0001,
+      "loss": 1.4887,
+      "step": 300
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5735897435897436,
+      "eval_loss": 1.6172032356262207,
+      "eval_runtime": 5.0425,
+      "eval_samples_per_second": 99.158,
+      "eval_steps_per_second": 12.494,
+      "step": 375
+    },
+    {
+      "epoch": 2.0,
+      "eval_exact_match": 19.8,
+      "eval_f1": 28.77095238095238,
+      "step": 375
+    },
+    {
+      "epoch": 2.1333333333333333,
+      "grad_norm": 1.28153395652771,
+      "learning_rate": 0.0001,
+      "loss": 1.4304,
+      "step": 400
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 1.3190444707870483,
+      "learning_rate": 0.0001,
+      "loss": 1.2346,
+      "step": 500
+    },
+    {
+      "epoch": 2.997333333333333,
+      "eval_accuracy": 0.5706153846153846,
+      "eval_loss": 1.6950472593307495,
+      "eval_runtime": 5.1981,
+      "eval_samples_per_second": 96.188,
+      "eval_steps_per_second": 12.12,
+      "step": 562
+    },
+    {
+      "epoch": 2.997333333333333,
+      "eval_exact_match": 19.4,
+      "eval_f1": 28.241904761904756,
+      "step": 562
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 1.5040416717529297,
+      "learning_rate": 0.0001,
+      "loss": 1.1583,
+      "step": 600
+    },
+    {
+      "epoch": 3.7333333333333334,
+      "grad_norm": 1.512290358543396,
+      "learning_rate": 0.0001,
+      "loss": 1.0104,
+      "step": 700
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5655897435897436,
+      "eval_loss": 1.7944799661636353,
+      "eval_runtime": 5.3441,
+      "eval_samples_per_second": 93.561,
+      "eval_steps_per_second": 11.789,
+      "step": 750
+    },
+    {
+      "epoch": 4.0,
+      "eval_exact_match": 15.8,
+      "eval_f1": 23.639523809523823,
+      "step": 750
+    },
+    {
+      "epoch": 4.266666666666667,
+      "grad_norm": 1.698282241821289,
+      "learning_rate": 0.0001,
+      "loss": 0.9275,
+      "step": 800
+    },
+    {
+      "epoch": 4.8,
+      "grad_norm": 1.6007401943206787,
+      "learning_rate": 0.0001,
+      "loss": 0.843,
+      "step": 900
+    },
+    {
+      "epoch": 4.997333333333334,
+      "eval_accuracy": 0.5632820512820513,
+      "eval_loss": 1.9219905138015747,
+      "eval_runtime": 5.2659,
+      "eval_samples_per_second": 94.951,
+      "eval_steps_per_second": 11.964,
+      "step": 937
+    },
+    {
+      "epoch": 4.997333333333334,
+      "eval_exact_match": 15.8,
+      "eval_f1": 24.631904761904778,
+      "step": 937
+    },
+    {
+      "epoch": 5.333333333333333,
+      "grad_norm": 1.5375592708587646,
+      "learning_rate": 0.0001,
+      "loss": 0.7516,
+      "step": 1000
+    },
+    {
+      "epoch": 5.866666666666667,
+      "grad_norm": 1.718989372253418,
+      "learning_rate": 0.0001,
+      "loss": 0.7199,
+      "step": 1100
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5616923076923077,
+      "eval_loss": 2.016322612762451,
+      "eval_runtime": 5.116,
+      "eval_samples_per_second": 97.734,
+      "eval_steps_per_second": 12.314,
+      "step": 1125
+    },
+    {
+      "epoch": 6.0,
+      "eval_exact_match": 14.4,
+      "eval_f1": 22.39285714285715,
+      "step": 1125
+    }
+  ],
+  "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": 2.059723965517005e+16,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-1125/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d667d5e0e40fe951f9fe0d2cd7fb4682a608b09af78a113a245307211adedfbc
+size 5304