Training in progress, epoch 6, checkpoint

checkpoint-1312/README.md

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

checkpoint-1312/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-1312/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3bbdd162e3be3fc5c38c789c03bc23576089a6d6fcac0be3329a5150064833c5
+size 112130416

checkpoint-1312/added_tokens.json

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

checkpoint-1312/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3825ff8b930f97c318d10a983bb60fde0f4874e49a5e765d3f0ff069c62bd9dc
+size 224537202

checkpoint-1312/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d894975494ee995790df6f1939a748282a440a9e2edf7b2ece08d18726db57b1
+size 15024

checkpoint-1312/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7f93abb626d9a0975072f556c075c9d2daa51720816ef62a77bd9bcac5ea0158
+size 15024

checkpoint-1312/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:23329dd8b199240fef9f8e8be2f6693c0269bb5933f8664d59192efbd076f1dd
+size 15024

checkpoint-1312/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2076f60c55543eb62840db8d4577b4b4492593f7ceead547ff0c13e920c1a588
+size 15024

checkpoint-1312/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:958dddeb11322a2f5484c651e2ee1ed82c5b95e61bb191f7a0726b85c1aa6329
+size 1064

checkpoint-1312/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-1312/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-1312/trainer_state.json

@@ -0,0 +1,257 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 6.997333333333334,
+  "eval_steps": 500,
+  "global_step": 1312,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.5333333333333333,
+      "grad_norm": 0.271308958530426,
+      "learning_rate": 5e-05,
+      "loss": 1.8262,
+      "step": 100
+    },
+    {
+      "epoch": 0.9973333333333333,
+      "eval_accuracy": 0.6047174887892377,
+      "eval_loss": 1.7027028799057007,
+      "eval_runtime": 6.6338,
+      "eval_samples_per_second": 75.372,
+      "eval_steps_per_second": 9.497,
+      "step": 187
+    },
+    {
+      "epoch": 0.9973333333333333,
+      "eval_exact_match": 14.6,
+      "eval_f1": 23.665714285714298,
+      "eval_qa_bleu": 10.24028390049623,
+      "eval_qa_exact_match": 0.114,
+      "eval_recite_bleu": 12.525534079250063,
+      "eval_recite_exact_match": 0.0,
+      "step": 187
+    },
+    {
+      "epoch": 1.0666666666666667,
+      "grad_norm": 0.301651269197464,
+      "learning_rate": 5e-05,
+      "loss": 1.7424,
+      "step": 200
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 0.35949045419692993,
+      "learning_rate": 5e-05,
+      "loss": 1.6978,
+      "step": 300
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6068699551569506,
+      "eval_loss": 1.6845285892486572,
+      "eval_runtime": 5.7524,
+      "eval_samples_per_second": 86.92,
+      "eval_steps_per_second": 10.952,
+      "step": 375
+    },
+    {
+      "epoch": 2.0,
+      "eval_exact_match": 14.6,
+      "eval_f1": 23.74190476190477,
+      "eval_qa_bleu": 9.70432698897927,
+      "eval_qa_exact_match": 0.114,
+      "eval_recite_bleu": 13.014822566317982,
+      "eval_recite_exact_match": 0.006,
+      "step": 375
+    },
+    {
+      "epoch": 2.1333333333333333,
+      "grad_norm": 0.41301965713500977,
+      "learning_rate": 5e-05,
+      "loss": 1.694,
+      "step": 400
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 0.5057485699653625,
+      "learning_rate": 5e-05,
+      "loss": 1.6324,
+      "step": 500
+    },
+    {
+      "epoch": 2.997333333333333,
+      "eval_accuracy": 0.6079551569506726,
+      "eval_loss": 1.6851012706756592,
+      "eval_runtime": 5.5314,
+      "eval_samples_per_second": 90.394,
+      "eval_steps_per_second": 11.39,
+      "step": 562
+    },
+    {
+      "epoch": 2.997333333333333,
+      "eval_exact_match": 14.2,
+      "eval_f1": 23.29190476190477,
+      "eval_qa_bleu": 10.724307586340888,
+      "eval_qa_exact_match": 0.104,
+      "eval_recite_bleu": 12.83531066630853,
+      "eval_recite_exact_match": 0.004,
+      "step": 562
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 0.6363334655761719,
+      "learning_rate": 5e-05,
+      "loss": 1.6063,
+      "step": 600
+    },
+    {
+      "epoch": 3.7333333333333334,
+      "grad_norm": 0.6466453671455383,
+      "learning_rate": 5e-05,
+      "loss": 1.5562,
+      "step": 700
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6071121076233184,
+      "eval_loss": 1.6969172954559326,
+      "eval_runtime": 6.2612,
+      "eval_samples_per_second": 79.857,
+      "eval_steps_per_second": 10.062,
+      "step": 750
+    },
+    {
+      "epoch": 4.0,
+      "eval_exact_match": 14.0,
+      "eval_f1": 21.34333333333334,
+      "eval_qa_bleu": 10.250340106635626,
+      "eval_qa_exact_match": 0.084,
+      "eval_recite_bleu": 12.741830585824024,
+      "eval_recite_exact_match": 0.004,
+      "step": 750
+    },
+    {
+      "epoch": 4.266666666666667,
+      "grad_norm": 0.8162885904312134,
+      "learning_rate": 5e-05,
+      "loss": 1.5052,
+      "step": 800
+    },
+    {
+      "epoch": 4.8,
+      "grad_norm": 0.8799439072608948,
+      "learning_rate": 5e-05,
+      "loss": 1.469,
+      "step": 900
+    },
+    {
+      "epoch": 4.997333333333334,
+      "eval_accuracy": 0.6059282511210762,
+      "eval_loss": 1.7216957807540894,
+      "eval_runtime": 6.3012,
+      "eval_samples_per_second": 79.35,
+      "eval_steps_per_second": 9.998,
+      "step": 937
+    },
+    {
+      "epoch": 4.997333333333334,
+      "eval_exact_match": 12.6,
+      "eval_f1": 21.041428571428575,
+      "eval_qa_bleu": 8.239984727774104,
+      "eval_qa_exact_match": 0.094,
+      "eval_recite_bleu": 12.327373697720192,
+      "eval_recite_exact_match": 0.002,
+      "step": 937
+    },
+    {
+      "epoch": 5.333333333333333,
+      "grad_norm": 0.9834449291229248,
+      "learning_rate": 5e-05,
+      "loss": 1.3845,
+      "step": 1000
+    },
+    {
+      "epoch": 5.866666666666667,
+      "grad_norm": 0.9377098679542542,
+      "learning_rate": 5e-05,
+      "loss": 1.3927,
+      "step": 1100
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.6047354260089686,
+      "eval_loss": 1.7564114332199097,
+      "eval_runtime": 6.3447,
+      "eval_samples_per_second": 78.806,
+      "eval_steps_per_second": 9.93,
+      "step": 1125
+    },
+    {
+      "epoch": 6.0,
+      "eval_exact_match": 13.4,
+      "eval_f1": 21.70317460317461,
+      "eval_qa_bleu": 10.522554916579404,
+      "eval_qa_exact_match": 0.104,
+      "eval_recite_bleu": 12.383604385812577,
+      "eval_recite_exact_match": 0.006,
+      "step": 1125
+    },
+    {
+      "epoch": 6.4,
+      "grad_norm": 1.1172072887420654,
+      "learning_rate": 5e-05,
+      "loss": 1.2924,
+      "step": 1200
+    },
+    {
+      "epoch": 6.933333333333334,
+      "grad_norm": 1.0980473756790161,
+      "learning_rate": 5e-05,
+      "loss": 1.286,
+      "step": 1300
+    },
+    {
+      "epoch": 6.997333333333334,
+      "eval_accuracy": 0.6026995515695067,
+      "eval_loss": 1.7977122068405151,
+      "eval_runtime": 5.9021,
+      "eval_samples_per_second": 84.716,
+      "eval_steps_per_second": 10.674,
+      "step": 1312
+    },
+    {
+      "epoch": 6.997333333333334,
+      "eval_exact_match": 12.0,
+      "eval_f1": 20.07142857142858,
+      "eval_qa_bleu": 9.89685444999411,
+      "eval_qa_exact_match": 0.102,
+      "eval_recite_bleu": 12.555118604776576,
+      "eval_recite_exact_match": 0.008,
+      "step": 1312
+    }
+  ],
+  "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": 1.3220581835014144e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-1312/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01f9f0dc44c57fb44afb03320f4d02bf48066009e3bbcd96d8e4377b0171c5a5
+size 5304