Training in progress, epoch 4, checkpoint

checkpoint-1366/README.md

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

checkpoint-1366/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-1366/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4204b0aab0774cc025ea680fea8ef8c7d506aee82154d3a4026cfdf5b17b1a7d
+size 112130416

checkpoint-1366/added_tokens.json

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

checkpoint-1366/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:839d1f3f8a1acfdcf12fa0e1fbc07ca1a60cb87294a68f3f6ececf79ae083957
+size 224537202

checkpoint-1366/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cdb8744b941d0fd06028ae90a21bfe23dfa15b8e26f74fcd17127e342ca2178a
+size 15024

checkpoint-1366/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6ee537c325a31c8896a520d7266a89cda41773d8bc15df36acc89316fc4f2c0e
+size 15024

checkpoint-1366/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eae19af8f58e89a59292a7177a1b38847595b1822da33db4047f032f41b24425
+size 15024

checkpoint-1366/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4e61c72e6fb408e454871870271ab6b9b031c12c951bd71a207f889a86210f58
+size 15024

checkpoint-1366/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:73b1f0ea5aa9bcd41f5ff294b225c6e4f66f8bbaea90adb73daa15529c463d90
+size 1064

checkpoint-1366/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-1366/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-1366/trainer_state.json

@@ -0,0 +1,188 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 4.0,
+  "eval_steps": 500,
+  "global_step": 1366,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.29282576866764276,
+      "grad_norm": 0.26115021109580994,
+      "learning_rate": 3e-05,
+      "loss": 2.0719,
+      "step": 100
+    },
+    {
+      "epoch": 0.5856515373352855,
+      "grad_norm": 0.2679530680179596,
+      "learning_rate": 3e-05,
+      "loss": 1.969,
+      "step": 200
+    },
+    {
+      "epoch": 0.8784773060029283,
+      "grad_norm": 0.2916224002838135,
+      "learning_rate": 3e-05,
+      "loss": 1.9686,
+      "step": 300
+    },
+    {
+      "epoch": 0.9985358711566618,
+      "eval_accuracy": 0.43743589743589745,
+      "eval_loss": 3.6918885707855225,
+      "eval_runtime": 5.2974,
+      "eval_samples_per_second": 94.385,
+      "eval_steps_per_second": 11.893,
+      "step": 341
+    },
+    {
+      "epoch": 0.9985358711566618,
+      "eval_bleu": 0.1527908371131025,
+      "eval_exact_match": 0.012,
+      "eval_prefix_exact_match": 0.026,
+      "step": 341
+    },
+    {
+      "epoch": 1.171303074670571,
+      "grad_norm": 0.3534347712993622,
+      "learning_rate": 3e-05,
+      "loss": 1.9603,
+      "step": 400
+    },
+    {
+      "epoch": 1.4641288433382138,
+      "grad_norm": 0.3338424563407898,
+      "learning_rate": 3e-05,
+      "loss": 1.9235,
+      "step": 500
+    },
+    {
+      "epoch": 1.7569546120058566,
+      "grad_norm": 0.4033390283584595,
+      "learning_rate": 3e-05,
+      "loss": 1.9337,
+      "step": 600
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.44764102564102565,
+      "eval_loss": 3.7484843730926514,
+      "eval_runtime": 5.0649,
+      "eval_samples_per_second": 98.718,
+      "eval_steps_per_second": 12.438,
+      "step": 683
+    },
+    {
+      "epoch": 2.0,
+      "eval_bleu": 0.21311912714932216,
+      "eval_exact_match": 0.02,
+      "eval_prefix_exact_match": 0.032,
+      "step": 683
+    },
+    {
+      "epoch": 2.049780380673499,
+      "grad_norm": 0.38882532715797424,
+      "learning_rate": 3e-05,
+      "loss": 1.9267,
+      "step": 700
+    },
+    {
+      "epoch": 2.342606149341142,
+      "grad_norm": 0.44310709834098816,
+      "learning_rate": 3e-05,
+      "loss": 1.8823,
+      "step": 800
+    },
+    {
+      "epoch": 2.6354319180087846,
+      "grad_norm": 0.5196680426597595,
+      "learning_rate": 3e-05,
+      "loss": 1.901,
+      "step": 900
+    },
+    {
+      "epoch": 2.9282576866764276,
+      "grad_norm": 0.5293437242507935,
+      "learning_rate": 3e-05,
+      "loss": 1.9033,
+      "step": 1000
+    },
+    {
+      "epoch": 2.998535871156662,
+      "eval_accuracy": 0.4495897435897436,
+      "eval_loss": 3.882551670074463,
+      "eval_runtime": 5.1314,
+      "eval_samples_per_second": 97.439,
+      "eval_steps_per_second": 12.277,
+      "step": 1024
+    },
+    {
+      "epoch": 2.998535871156662,
+      "eval_bleu": 0.24307488023362148,
+      "eval_exact_match": 0.016,
+      "eval_prefix_exact_match": 0.032,
+      "step": 1024
+    },
+    {
+      "epoch": 3.22108345534407,
+      "grad_norm": 0.552943229675293,
+      "learning_rate": 3e-05,
+      "loss": 1.8529,
+      "step": 1100
+    },
+    {
+      "epoch": 3.513909224011713,
+      "grad_norm": 0.6421984434127808,
+      "learning_rate": 3e-05,
+      "loss": 1.8342,
+      "step": 1200
+    },
+    {
+      "epoch": 3.8067349926793557,
+      "grad_norm": 0.7155198454856873,
+      "learning_rate": 3e-05,
+      "loss": 1.857,
+      "step": 1300
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.4481025641025641,
+      "eval_loss": 3.970107316970825,
+      "eval_runtime": 5.1563,
+      "eval_samples_per_second": 96.969,
+      "eval_steps_per_second": 12.218,
+      "step": 1366
+    },
+    {
+      "epoch": 4.0,
+      "eval_bleu": 0.25309406645138827,
+      "eval_exact_match": 0.016,
+      "eval_prefix_exact_match": 0.044,
+      "step": 1366
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 17050,
+  "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": 1.1730026735914189e+17,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-1366/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f9cef08c7f10bf396c269dfc26104fd5a94756f5431d6448312805ef8e298380
+size 5304