Training in progress, epoch 8, checkpoint

checkpoint-2000/README.md

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

checkpoint-2000/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-2000/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5323717e0c834f99ec7bef5335dec1d3b09c99f479e8a03d881b6e8ff4839c0e
+size 112130416

checkpoint-2000/added_tokens.json

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

checkpoint-2000/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e8771de61b2fac7afcc93457bdff00eb95687d8745e17faf09f697879f1c8281
+size 224537202

checkpoint-2000/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:07c1370a06ec1f53023e2416c33a2293ef395ecf032291af18bb0c5b1eae92b6
+size 15024

checkpoint-2000/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:103039b346f6ec1d223901d0d74dfeea18bd5b544e923d159675f90d68b234af
+size 15024

checkpoint-2000/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c44078ae73004cbb8880e105a4e47a0fc1413a2a1f75c6e1165be4a43dec1a84
+size 15024

checkpoint-2000/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7d7b219e00fbf3bca6fe3bbbc9d7307f0ecf65b7d53fe0ff4b2eb5f32993e4b
+size 15024

checkpoint-2000/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e91519ad666af6acc4beafd44aa617aefb90f632388a0f75cbf88f957400c190
+size 1064

checkpoint-2000/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-2000/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-2000/trainer_state.json

@@ -0,0 +1,293 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 8.0,
+  "eval_steps": 500,
+  "global_step": 2000,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.48472875356674194,
+      "learning_rate": 0.0003,
+      "loss": 2.3219,
+      "step": 100
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.4583953619003296,
+      "learning_rate": 0.0003,
+      "loss": 2.2487,
+      "step": 200
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5168253968253969,
+      "eval_loss": 2.311781167984009,
+      "eval_runtime": 5.4425,
+      "eval_samples_per_second": 91.87,
+      "eval_steps_per_second": 11.576,
+      "step": 250
+    },
+    {
+      "epoch": 1.0,
+      "eval_exact_match": 18.6,
+      "eval_f1": 26.999121042356347,
+      "step": 250
+    },
+    {
+      "epoch": 1.2,
+      "grad_norm": 0.6610094904899597,
+      "learning_rate": 0.0003,
+      "loss": 2.083,
+      "step": 300
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 0.7361629605293274,
+      "learning_rate": 0.0003,
+      "loss": 1.9274,
+      "step": 400
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.6793610453605652,
+      "learning_rate": 0.0003,
+      "loss": 1.9113,
+      "step": 500
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5153650793650794,
+      "eval_loss": 2.3749232292175293,
+      "eval_runtime": 5.2167,
+      "eval_samples_per_second": 95.846,
+      "eval_steps_per_second": 12.077,
+      "step": 500
+    },
+    {
+      "epoch": 2.0,
+      "eval_exact_match": 18.2,
+      "eval_f1": 26.209704184704197,
+      "step": 500
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 0.8843489289283752,
+      "learning_rate": 0.0003,
+      "loss": 1.4602,
+      "step": 600
+    },
+    {
+      "epoch": 2.8,
+      "grad_norm": 0.8380613923072815,
+      "learning_rate": 0.0003,
+      "loss": 1.4924,
+      "step": 700
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5098730158730159,
+      "eval_loss": 2.5289740562438965,
+      "eval_runtime": 5.3631,
+      "eval_samples_per_second": 93.23,
+      "eval_steps_per_second": 11.747,
+      "step": 750
+    },
+    {
+      "epoch": 3.0,
+      "eval_exact_match": 16.6,
+      "eval_f1": 24.526153291153296,
+      "step": 750
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 1.068247675895691,
+      "learning_rate": 0.0003,
+      "loss": 1.2869,
+      "step": 800
+    },
+    {
+      "epoch": 3.6,
+      "grad_norm": 1.041741967201233,
+      "learning_rate": 0.0003,
+      "loss": 1.0749,
+      "step": 900
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.267968773841858,
+      "learning_rate": 0.0003,
+      "loss": 1.1157,
+      "step": 1000
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5038412698412699,
+      "eval_loss": 2.794942855834961,
+      "eval_runtime": 5.0088,
+      "eval_samples_per_second": 99.825,
+      "eval_steps_per_second": 12.578,
+      "step": 1000
+    },
+    {
+      "epoch": 4.0,
+      "eval_exact_match": 16.4,
+      "eval_f1": 23.978561993561993,
+      "step": 1000
+    },
+    {
+      "epoch": 4.4,
+      "grad_norm": 1.1971005201339722,
+      "learning_rate": 0.0003,
+      "loss": 0.7367,
+      "step": 1100
+    },
+    {
+      "epoch": 4.8,
+      "grad_norm": 1.2383127212524414,
+      "learning_rate": 0.0003,
+      "loss": 0.7814,
+      "step": 1200
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.49853968253968256,
+      "eval_loss": 3.0137181282043457,
+      "eval_runtime": 5.1368,
+      "eval_samples_per_second": 97.337,
+      "eval_steps_per_second": 12.264,
+      "step": 1250
+    },
+    {
+      "epoch": 5.0,
+      "eval_exact_match": 14.8,
+      "eval_f1": 21.880294705294705,
+      "step": 1250
+    },
+    {
+      "epoch": 5.2,
+      "grad_norm": 0.9558005928993225,
+      "learning_rate": 0.0003,
+      "loss": 0.6786,
+      "step": 1300
+    },
+    {
+      "epoch": 5.6,
+      "grad_norm": 1.1341361999511719,
+      "learning_rate": 0.0003,
+      "loss": 0.5741,
+      "step": 1400
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 1.1375033855438232,
+      "learning_rate": 0.0003,
+      "loss": 0.605,
+      "step": 1500
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.49657142857142855,
+      "eval_loss": 3.220507860183716,
+      "eval_runtime": 5.2708,
+      "eval_samples_per_second": 94.863,
+      "eval_steps_per_second": 11.953,
+      "step": 1500
+    },
+    {
+      "epoch": 6.0,
+      "eval_exact_match": 15.4,
+      "eval_f1": 22.281807987334307,
+      "step": 1500
+    },
+    {
+      "epoch": 6.4,
+      "grad_norm": 1.118528962135315,
+      "learning_rate": 0.0003,
+      "loss": 0.4684,
+      "step": 1600
+    },
+    {
+      "epoch": 6.8,
+      "grad_norm": 0.9311291575431824,
+      "learning_rate": 0.0003,
+      "loss": 0.4968,
+      "step": 1700
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.4946984126984127,
+      "eval_loss": 3.3760199546813965,
+      "eval_runtime": 5.0734,
+      "eval_samples_per_second": 98.554,
+      "eval_steps_per_second": 12.418,
+      "step": 1750
+    },
+    {
+      "epoch": 7.0,
+      "eval_exact_match": 14.8,
+      "eval_f1": 21.79480811586075,
+      "step": 1750
+    },
+    {
+      "epoch": 7.2,
+      "grad_norm": 1.0299904346466064,
+      "learning_rate": 0.0003,
+      "loss": 0.4631,
+      "step": 1800
+    },
+    {
+      "epoch": 7.6,
+      "grad_norm": 0.7518107295036316,
+      "learning_rate": 0.0003,
+      "loss": 0.4411,
+      "step": 1900
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.1140515804290771,
+      "learning_rate": 0.0003,
+      "loss": 0.4597,
+      "step": 2000
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.49234920634920637,
+      "eval_loss": 3.4634218215942383,
+      "eval_runtime": 5.8652,
+      "eval_samples_per_second": 85.249,
+      "eval_steps_per_second": 10.741,
+      "step": 2000
+    },
+    {
+      "epoch": 8.0,
+      "eval_exact_match": 15.0,
+      "eval_f1": 21.617115196631318,
+      "step": 2000
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 12500,
+  "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": 5.174154322404966e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-2000/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dae73bb50794744b259c1eaaeb67f6ae0b1d0f39b0a32b5f1162fc82e3a5d6a6
+size 5304