Training in progress, epoch 6, checkpoint

checkpoint-3174/README.md

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

checkpoint-3174/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-3174/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a91bc802eeccfdd5406868a40f61596dd728cfb81e313f0ed1a7e95857d5f3c7
+size 112130416

checkpoint-3174/added_tokens.json

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

checkpoint-3174/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d3f3acddb45d472750dadd5bcc8d02743a0de4001f2ea23625de30eb4b3b529
+size 224537202

checkpoint-3174/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dc863aa478b9bf3fa376185744303c081b242ba0b2ae60d5244c2838c8733dc1
+size 15024

checkpoint-3174/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:39c9dac0002f10ef59eb87dd5941bff107341df1dd6dcf7b447b6b1454d35d43
+size 15024

checkpoint-3174/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9e3f302a6043bc917d7ad25b3e8430e6d723ddc28c88ede52db03beb746fbcca
+size 15024

checkpoint-3174/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6b4d07cd7635cf4f76379bd10e07cc65b9759b990c35c070acda6abcff519bab
+size 15024

checkpoint-3174/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:48fade52b55558f869e0c0e9de0ff551240a3acd6e2aef7f8087d4c58f60415c
+size 1064

checkpoint-3174/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-3174/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-3174/trainer_state.json

@@ -0,0 +1,340 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 6.0,
+  "eval_steps": 500,
+  "global_step": 3174,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.1890359168241966,
+      "grad_norm": 0.5558584332466125,
+      "learning_rate": 3e-05,
+      "loss": 2.1103,
+      "step": 100
+    },
+    {
+      "epoch": 0.3780718336483932,
+      "grad_norm": 0.4733140468597412,
+      "learning_rate": 3e-05,
+      "loss": 1.931,
+      "step": 200
+    },
+    {
+      "epoch": 0.5671077504725898,
+      "grad_norm": 0.48580050468444824,
+      "learning_rate": 3e-05,
+      "loss": 1.8948,
+      "step": 300
+    },
+    {
+      "epoch": 0.7561436672967864,
+      "grad_norm": 0.40655234456062317,
+      "learning_rate": 3e-05,
+      "loss": 1.8988,
+      "step": 400
+    },
+    {
+      "epoch": 0.945179584120983,
+      "grad_norm": 0.6303955316543579,
+      "learning_rate": 3e-05,
+      "loss": 1.8745,
+      "step": 500
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5701538461538461,
+      "eval_loss": 1.6455246210098267,
+      "eval_runtime": 5.0805,
+      "eval_samples_per_second": 98.415,
+      "eval_steps_per_second": 12.4,
+      "step": 529
+    },
+    {
+      "epoch": 1.0,
+      "eval_exact_match": 15.0,
+      "eval_f1": 23.927619047619068,
+      "step": 529
+    },
+    {
+      "epoch": 1.1342155009451795,
+      "grad_norm": 0.4278818666934967,
+      "learning_rate": 3e-05,
+      "loss": 1.8798,
+      "step": 600
+    },
+    {
+      "epoch": 1.3232514177693762,
+      "grad_norm": 0.449830025434494,
+      "learning_rate": 3e-05,
+      "loss": 1.8613,
+      "step": 700
+    },
+    {
+      "epoch": 1.5122873345935728,
+      "grad_norm": 0.5130725502967834,
+      "learning_rate": 3e-05,
+      "loss": 1.8709,
+      "step": 800
+    },
+    {
+      "epoch": 1.7013232514177694,
+      "grad_norm": 0.4889715611934662,
+      "learning_rate": 3e-05,
+      "loss": 1.8448,
+      "step": 900
+    },
+    {
+      "epoch": 1.8903591682419658,
+      "grad_norm": 11.186380386352539,
+      "learning_rate": 3e-05,
+      "loss": 1.8764,
+      "step": 1000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5723589743589743,
+      "eval_loss": 1.6143561601638794,
+      "eval_runtime": 5.0466,
+      "eval_samples_per_second": 99.077,
+      "eval_steps_per_second": 12.484,
+      "step": 1058
+    },
+    {
+      "epoch": 2.0,
+      "eval_exact_match": 15.6,
+      "eval_f1": 25.825238095238113,
+      "step": 1058
+    },
+    {
+      "epoch": 2.0793950850661624,
+      "grad_norm": 0.5414931774139404,
+      "learning_rate": 3e-05,
+      "loss": 1.8398,
+      "step": 1100
+    },
+    {
+      "epoch": 2.268431001890359,
+      "grad_norm": 7.441005706787109,
+      "learning_rate": 3e-05,
+      "loss": 1.8348,
+      "step": 1200
+    },
+    {
+      "epoch": 2.4574669187145557,
+      "grad_norm": 24.312145233154297,
+      "learning_rate": 3e-05,
+      "loss": 1.8421,
+      "step": 1300
+    },
+    {
+      "epoch": 2.6465028355387523,
+      "grad_norm": 0.7407662868499756,
+      "learning_rate": 3e-05,
+      "loss": 1.8292,
+      "step": 1400
+    },
+    {
+      "epoch": 2.835538752362949,
+      "grad_norm": 0.9653642177581787,
+      "learning_rate": 3e-05,
+      "loss": 1.828,
+      "step": 1500
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5734358974358974,
+      "eval_loss": 1.617864727973938,
+      "eval_runtime": 5.2153,
+      "eval_samples_per_second": 95.872,
+      "eval_steps_per_second": 12.08,
+      "step": 1587
+    },
+    {
+      "epoch": 3.0,
+      "eval_exact_match": 16.2,
+      "eval_f1": 25.811428571428586,
+      "step": 1587
+    },
+    {
+      "epoch": 3.0245746691871456,
+      "grad_norm": 450.6055603027344,
+      "learning_rate": 3e-05,
+      "loss": 1.8204,
+      "step": 1600
+    },
+    {
+      "epoch": 3.213610586011342,
+      "grad_norm": 2.1127028465270996,
+      "learning_rate": 3e-05,
+      "loss": 1.8361,
+      "step": 1700
+    },
+    {
+      "epoch": 3.402646502835539,
+      "grad_norm": 564.9320068359375,
+      "learning_rate": 3e-05,
+      "loss": 1.8338,
+      "step": 1800
+    },
+    {
+      "epoch": 3.5916824196597354,
+      "grad_norm": 52.5078010559082,
+      "learning_rate": 3e-05,
+      "loss": 1.8075,
+      "step": 1900
+    },
+    {
+      "epoch": 3.780718336483932,
+      "grad_norm": 27.87557601928711,
+      "learning_rate": 3e-05,
+      "loss": 1.8175,
+      "step": 2000
+    },
+    {
+      "epoch": 3.9697542533081287,
+      "grad_norm": 4.810656547546387,
+      "learning_rate": 3e-05,
+      "loss": 1.8218,
+      "step": 2100
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5728205128205128,
+      "eval_loss": 1.6224406957626343,
+      "eval_runtime": 5.0588,
+      "eval_samples_per_second": 98.838,
+      "eval_steps_per_second": 12.454,
+      "step": 2116
+    },
+    {
+      "epoch": 4.0,
+      "eval_exact_match": 17.4,
+      "eval_f1": 27.194285714285726,
+      "step": 2116
+    },
+    {
+      "epoch": 4.158790170132325,
+      "grad_norm": 13.004076957702637,
+      "learning_rate": 3e-05,
+      "loss": 1.8089,
+      "step": 2200
+    },
+    {
+      "epoch": 4.3478260869565215,
+      "grad_norm": 2.23368501663208,
+      "learning_rate": 3e-05,
+      "loss": 1.8292,
+      "step": 2300
+    },
+    {
+      "epoch": 4.536862003780718,
+      "grad_norm": 0.8635586500167847,
+      "learning_rate": 3e-05,
+      "loss": 1.7992,
+      "step": 2400
+    },
+    {
+      "epoch": 4.725897920604915,
+      "grad_norm": 6.5326738357543945,
+      "learning_rate": 3e-05,
+      "loss": 1.7955,
+      "step": 2500
+    },
+    {
+      "epoch": 4.914933837429111,
+      "grad_norm": 463.4647521972656,
+      "learning_rate": 3e-05,
+      "loss": 1.7953,
+      "step": 2600
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5740512820512821,
+      "eval_loss": 1.6217337846755981,
+      "eval_runtime": 5.1898,
+      "eval_samples_per_second": 96.342,
+      "eval_steps_per_second": 12.139,
+      "step": 2645
+    },
+    {
+      "epoch": 5.0,
+      "eval_exact_match": 17.6,
+      "eval_f1": 26.161428571428587,
+      "step": 2645
+    },
+    {
+      "epoch": 5.103969754253308,
+      "grad_norm": 203.82815551757812,
+      "learning_rate": 3e-05,
+      "loss": 1.7845,
+      "step": 2700
+    },
+    {
+      "epoch": 5.293005671077505,
+      "grad_norm": 660.8076782226562,
+      "learning_rate": 3e-05,
+      "loss": 1.7909,
+      "step": 2800
+    },
+    {
+      "epoch": 5.482041587901701,
+      "grad_norm": 5727.51806640625,
+      "learning_rate": 3e-05,
+      "loss": 1.7909,
+      "step": 2900
+    },
+    {
+      "epoch": 5.671077504725898,
+      "grad_norm": 4880698.0,
+      "learning_rate": 3e-05,
+      "loss": 3.5781,
+      "step": 3000
+    },
+    {
+      "epoch": 5.8601134215500945,
+      "grad_norm": 1192.6517333984375,
+      "learning_rate": 3e-05,
+      "loss": 3.8703,
+      "step": 3100
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5433846153846154,
+      "eval_loss": 1.886108636856079,
+      "eval_runtime": 5.5162,
+      "eval_samples_per_second": 90.643,
+      "eval_steps_per_second": 11.421,
+      "step": 3174
+    },
+    {
+      "epoch": 6.0,
+      "eval_exact_match": 11.4,
+      "eval_f1": 20.358571428571437,
+      "step": 3174
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 26450,
+  "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": 2.5829249149986406e+17,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-3174/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3895667ea94410ecd283430024ce6f8557816c9d86efd03783de342ef988d79c
+size 5304