Training in progress, epoch 8, checkpoint

checkpoint-1500/README.md

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

checkpoint-1500/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-1500/adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d6fcc35a21095e7e1c9b579defcca479046cd817e8311c2a60977811e096cb2
+size 112130416

checkpoint-1500/added_tokens.json

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

checkpoint-1500/optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6f957034fa15e40e4f9c36b6eb861662d7c81429d246bb9ca8da7712cbe01e13
+size 224537202

checkpoint-1500/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3f9f36cd83b31c4a9620c6d5509f917c9719fac614d4e0ddabd439dc283a1c2f
+size 15024

checkpoint-1500/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d5866cb3d7d81e6ad54d5d26c168c433ef28516b45e50c1acf3abfb56dfb9c9d
+size 15024

checkpoint-1500/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5eb3db160a8df9a37091a49e50e372ea835e1456a39b7f4f221bead9e6ec0b16
+size 15024

checkpoint-1500/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bba92bec10a89fa1b0e0170c385edc91e5f526f5568566995dc5428d51907cf7
+size 15024

checkpoint-1500/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f226c38a637eacea5ba36b19dd24a013288ce8b2f58c6f7ea9303ec1896824b9
+size 1064

checkpoint-1500/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-1500/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-1500/trainer_state.json

@@ -0,0 +1,258 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 8.0,
+  "eval_steps": 500,
+  "global_step": 1500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.5333333333333333,
+      "grad_norm": 0.9283596873283386,
+      "learning_rate": 3e-05,
+      "loss": 1.963,
+      "step": 100
+    },
+    {
+      "epoch": 0.9973333333333333,
+      "eval_accuracy": 0.5695384615384615,
+      "eval_loss": 1.6438876390457153,
+      "eval_runtime": 5.0534,
+      "eval_samples_per_second": 98.943,
+      "eval_steps_per_second": 12.467,
+      "step": 187
+    },
+    {
+      "epoch": 0.9973333333333333,
+      "eval_exact_match": 16.2,
+      "eval_f1": 25.888571428571442,
+      "step": 187
+    },
+    {
+      "epoch": 1.0666666666666667,
+      "grad_norm": 1.0407154560089111,
+      "learning_rate": 3e-05,
+      "loss": 1.6668,
+      "step": 200
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 1.1281720399856567,
+      "learning_rate": 3e-05,
+      "loss": 1.6099,
+      "step": 300
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5733333333333334,
+      "eval_loss": 1.618270754814148,
+      "eval_runtime": 5.1983,
+      "eval_samples_per_second": 96.186,
+      "eval_steps_per_second": 12.119,
+      "step": 375
+    },
+    {
+      "epoch": 2.0,
+      "eval_exact_match": 16.6,
+      "eval_f1": 26.287619047619057,
+      "step": 375
+    },
+    {
+      "epoch": 2.1333333333333333,
+      "grad_norm": 1.2508225440979004,
+      "learning_rate": 3e-05,
+      "loss": 1.5866,
+      "step": 400
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 1.3197834491729736,
+      "learning_rate": 3e-05,
+      "loss": 1.524,
+      "step": 500
+    },
+    {
+      "epoch": 2.997333333333333,
+      "eval_accuracy": 0.5743589743589743,
+      "eval_loss": 1.6164497137069702,
+      "eval_runtime": 5.0495,
+      "eval_samples_per_second": 99.019,
+      "eval_steps_per_second": 12.476,
+      "step": 562
+    },
+    {
+      "epoch": 2.997333333333333,
+      "eval_exact_match": 17.8,
+      "eval_f1": 27.35428571428571,
+      "step": 562
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 1.6320394277572632,
+      "learning_rate": 3e-05,
+      "loss": 1.4752,
+      "step": 600
+    },
+    {
+      "epoch": 3.7333333333333334,
+      "grad_norm": 1.7101725339889526,
+      "learning_rate": 3e-05,
+      "loss": 1.3938,
+      "step": 700
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5728717948717948,
+      "eval_loss": 1.6376241445541382,
+      "eval_runtime": 4.9662,
+      "eval_samples_per_second": 100.68,
+      "eval_steps_per_second": 12.686,
+      "step": 750
+    },
+    {
+      "epoch": 4.0,
+      "eval_exact_match": 18.0,
+      "eval_f1": 27.498571428571427,
+      "step": 750
+    },
+    {
+      "epoch": 4.266666666666667,
+      "grad_norm": 2.1727664470672607,
+      "learning_rate": 3e-05,
+      "loss": 1.342,
+      "step": 800
+    },
+    {
+      "epoch": 4.8,
+      "grad_norm": 2.341384172439575,
+      "learning_rate": 3e-05,
+      "loss": 1.2685,
+      "step": 900
+    },
+    {
+      "epoch": 4.997333333333334,
+      "eval_accuracy": 0.5699487179487179,
+      "eval_loss": 1.684618353843689,
+      "eval_runtime": 5.5457,
+      "eval_samples_per_second": 90.16,
+      "eval_steps_per_second": 11.36,
+      "step": 937
+    },
+    {
+      "epoch": 4.997333333333334,
+      "eval_exact_match": 19.2,
+      "eval_f1": 28.190952380952375,
+      "step": 937
+    },
+    {
+      "epoch": 5.333333333333333,
+      "grad_norm": 2.5624172687530518,
+      "learning_rate": 3e-05,
+      "loss": 1.1847,
+      "step": 1000
+    },
+    {
+      "epoch": 5.866666666666667,
+      "grad_norm": 2.8159019947052,
+      "learning_rate": 3e-05,
+      "loss": 1.1591,
+      "step": 1100
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5672820512820513,
+      "eval_loss": 1.7546838521957397,
+      "eval_runtime": 5.0186,
+      "eval_samples_per_second": 99.629,
+      "eval_steps_per_second": 12.553,
+      "step": 1125
+    },
+    {
+      "epoch": 6.0,
+      "eval_exact_match": 16.6,
+      "eval_f1": 25.054761904761914,
+      "step": 1125
+    },
+    {
+      "epoch": 6.4,
+      "grad_norm": 3.056166410446167,
+      "learning_rate": 3e-05,
+      "loss": 1.0486,
+      "step": 1200
+    },
+    {
+      "epoch": 6.933333333333334,
+      "grad_norm": 2.873281240463257,
+      "learning_rate": 3e-05,
+      "loss": 1.0444,
+      "step": 1300
+    },
+    {
+      "epoch": 6.997333333333334,
+      "eval_accuracy": 0.5643076923076923,
+      "eval_loss": 1.8394943475723267,
+      "eval_runtime": 5.5931,
+      "eval_samples_per_second": 89.396,
+      "eval_steps_per_second": 11.264,
+      "step": 1312
+    },
+    {
+      "epoch": 6.997333333333334,
+      "eval_exact_match": 15.8,
+      "eval_f1": 23.925714285714296,
+      "step": 1312
+    },
+    {
+      "epoch": 7.466666666666667,
+      "grad_norm": 3.069322109222412,
+      "learning_rate": 3e-05,
+      "loss": 0.9374,
+      "step": 1400
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 3.2732129096984863,
+      "learning_rate": 3e-05,
+      "loss": 0.9535,
+      "step": 1500
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5612820512820513,
+      "eval_loss": 1.9008216857910156,
+      "eval_runtime": 5.1357,
+      "eval_samples_per_second": 97.358,
+      "eval_steps_per_second": 12.267,
+      "step": 1500
+    },
+    {
+      "epoch": 8.0,
+      "eval_exact_match": 15.2,
+      "eval_f1": 23.12063492063493,
+      "step": 1500
+    }
+  ],
+  "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": 2.746516824601395e+16,
+  "train_batch_size": 2,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-1500/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1a175de730fa6917b3a611bb9e53766640b7a5e4a426f825f2a602cf98661e47
+size 5304