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:d45628da3f4de2eca52afc4aa79f363708a45b7c4a4b93c4e1899925a2904df1
+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:bcef7189c2c78ba90bed6bca61b3bfe903b7d2985e69e56a6c71e2ed2e317623
+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 sha256:41fa2d63ad9d30108b6fe47b5f1e94cfa7b6c8fc0f0bbd911dff3afe84de05d3
+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": 0.3393945097923279,
+      "learning_rate": 0.0003,
+      "loss": 1.951,
+      "step": 100
+    },
+    {
+      "epoch": 0.3780718336483932,
+      "grad_norm": 8017.20361328125,
+      "learning_rate": 0.0003,
+      "loss": 2.0285,
+      "step": 200
+    },
+    {
+      "epoch": 0.5671077504725898,
+      "grad_norm": 9146.7236328125,
+      "learning_rate": 0.0003,
+      "loss": 6.5404,
+      "step": 300
+    },
+    {
+      "epoch": 0.7561436672967864,
+      "grad_norm": 574.934326171875,
+      "learning_rate": 0.0003,
+      "loss": 7.7357,
+      "step": 400
+    },
+    {
+      "epoch": 0.945179584120983,
+      "grad_norm": 25128.55859375,
+      "learning_rate": 0.0003,
+      "loss": 6.3771,
+      "step": 500
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.24974358974358973,
+      "eval_loss": 10.19681453704834,
+      "eval_runtime": 5.1408,
+      "eval_samples_per_second": 97.26,
+      "eval_steps_per_second": 12.255,
+      "step": 529
+    },
+    {
+      "epoch": 1.0,
+      "eval_exact_match": 0.0,
+      "eval_f1": 0.0,
+      "step": 529
+    },
+    {
+      "epoch": 1.1342155009451795,
+      "grad_norm": 1332.8885498046875,
+      "learning_rate": 0.0003,
+      "loss": 11.0858,
+      "step": 600
+    },
+    {
+      "epoch": 1.3232514177693762,
+      "grad_norm": 3666.881591796875,
+      "learning_rate": 0.0003,
+      "loss": 9.483,
+      "step": 700
+    },
+    {
+      "epoch": 1.5122873345935728,
+      "grad_norm": 356847.0,
+      "learning_rate": 0.0003,
+      "loss": 8.069,
+      "step": 800
+    },
+    {
+      "epoch": 1.7013232514177694,
+      "grad_norm": 1061.877685546875,
+      "learning_rate": 0.0003,
+      "loss": 8.7493,
+      "step": 900
+    },
+    {
+      "epoch": 1.8903591682419658,
+      "grad_norm": 46670.3046875,
+      "learning_rate": 0.0003,
+      "loss": 8.3415,
+      "step": 1000
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.20507692307692307,
+      "eval_loss": 7.504091262817383,
+      "eval_runtime": 5.2437,
+      "eval_samples_per_second": 95.352,
+      "eval_steps_per_second": 12.014,
+      "step": 1058
+    },
+    {
+      "epoch": 2.0,
+      "eval_exact_match": 0.0,
+      "eval_f1": 0.0,
+      "step": 1058
+    },
+    {
+      "epoch": 2.0793950850661624,
+      "grad_norm": 2.0727381706237793,
+      "learning_rate": 0.0003,
+      "loss": 8.3511,
+      "step": 1100
+    },
+    {
+      "epoch": 2.268431001890359,
+      "grad_norm": 1.8185893297195435,
+      "learning_rate": 0.0003,
+      "loss": 7.3065,
+      "step": 1200
+    },
+    {
+      "epoch": 2.4574669187145557,
+      "grad_norm": 1620.93701171875,
+      "learning_rate": 0.0003,
+      "loss": 7.1302,
+      "step": 1300
+    },
+    {
+      "epoch": 2.6465028355387523,
+      "grad_norm": 1.0691380500793457,
+      "learning_rate": 0.0003,
+      "loss": 7.4423,
+      "step": 1400
+    },
+    {
+      "epoch": 2.835538752362949,
+      "grad_norm": 1.0605100393295288,
+      "learning_rate": 0.0003,
+      "loss": 7.6715,
+      "step": 1500
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.28507692307692306,
+      "eval_loss": 7.084368705749512,
+      "eval_runtime": 5.493,
+      "eval_samples_per_second": 91.026,
+      "eval_steps_per_second": 11.469,
+      "step": 1587
+    },
+    {
+      "epoch": 3.0,
+      "eval_exact_match": 0.0,
+      "eval_f1": 0.08,
+      "step": 1587
+    },
+    {
+      "epoch": 3.0245746691871456,
+      "grad_norm": 1967676800.0,
+      "learning_rate": 0.0003,
+      "loss": 7.8617,
+      "step": 1600
+    },
+    {
+      "epoch": 3.213610586011342,
+      "grad_norm": 83918778793984.0,
+      "learning_rate": 0.0003,
+      "loss": 8.083,
+      "step": 1700
+    },
+    {
+      "epoch": 3.402646502835539,
+      "grad_norm": 12349449.0,
+      "learning_rate": 0.0003,
+      "loss": 8.0382,
+      "step": 1800
+    },
+    {
+      "epoch": 3.5916824196597354,
+      "grad_norm": 52939337728.0,
+      "learning_rate": 0.0003,
+      "loss": 7.8761,
+      "step": 1900
+    },
+    {
+      "epoch": 3.780718336483932,
+      "grad_norm": 1381182.25,
+      "learning_rate": 0.0003,
+      "loss": 7.4665,
+      "step": 2000
+    },
+    {
+      "epoch": 3.9697542533081287,
+      "grad_norm": 105.80550384521484,
+      "learning_rate": 0.0003,
+      "loss": 7.8182,
+      "step": 2100
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.23933333333333334,
+      "eval_loss": 7.05845308303833,
+      "eval_runtime": 5.0891,
+      "eval_samples_per_second": 98.248,
+      "eval_steps_per_second": 12.379,
+      "step": 2116
+    },
+    {
+      "epoch": 4.0,
+      "eval_exact_match": 0.0,
+      "eval_f1": 0.0,
+      "step": 2116
+    },
+    {
+      "epoch": 4.158790170132325,
+      "grad_norm": 16.110862731933594,
+      "learning_rate": 0.0003,
+      "loss": 7.9728,
+      "step": 2200
+    },
+    {
+      "epoch": 4.3478260869565215,
+      "grad_norm": 4.362427234649658,
+      "learning_rate": 0.0003,
+      "loss": 7.2875,
+      "step": 2300
+    },
+    {
+      "epoch": 4.536862003780718,
+      "grad_norm": 18039.095703125,
+      "learning_rate": 0.0003,
+      "loss": 7.3972,
+      "step": 2400
+    },
+    {
+      "epoch": 4.725897920604915,
+      "grad_norm": 2.3812594051160474e+17,
+      "learning_rate": 0.0003,
+      "loss": 7.4631,
+      "step": 2500
+    },
+    {
+      "epoch": 4.914933837429111,
+      "grad_norm": 3.9549436569213867,
+      "learning_rate": 0.0003,
+      "loss": 7.3322,
+      "step": 2600
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.22430769230769232,
+      "eval_loss": 6.72421932220459,
+      "eval_runtime": 5.1689,
+      "eval_samples_per_second": 96.731,
+      "eval_steps_per_second": 12.188,
+      "step": 2645
+    },
+    {
+      "epoch": 5.0,
+      "eval_exact_match": 0.0,
+      "eval_f1": 0.0,
+      "step": 2645
+    },
+    {
+      "epoch": 5.103969754253308,
+      "grad_norm": 120114.2421875,
+      "learning_rate": 0.0003,
+      "loss": 7.2428,
+      "step": 2700
+    },
+    {
+      "epoch": 5.293005671077505,
+      "grad_norm": 391.66119384765625,
+      "learning_rate": 0.0003,
+      "loss": 7.2495,
+      "step": 2800
+    },
+    {
+      "epoch": 5.482041587901701,
+      "grad_norm": 599371136.0,
+      "learning_rate": 0.0003,
+      "loss": 7.5915,
+      "step": 2900
+    },
+    {
+      "epoch": 5.671077504725898,
+      "grad_norm": 5585.97314453125,
+      "learning_rate": 0.0003,
+      "loss": 7.3014,
+      "step": 3000
+    },
+    {
+      "epoch": 5.8601134215500945,
+      "grad_norm": 268239548448768.0,
+      "learning_rate": 0.0003,
+      "loss": 7.259,
+      "step": 3100
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.24984615384615386,
+      "eval_loss": 6.531904220581055,
+      "eval_runtime": 5.0344,
+      "eval_samples_per_second": 99.317,
+      "eval_steps_per_second": 12.514,
+      "step": 3174
+    },
+    {
+      "epoch": 6.0,
+      "eval_exact_match": 0.0,
+      "eval_f1": 0.0,
+      "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 sha256:186f3ae462881ca0bb3a6b5295a97a245e0a78172be9c14948a2d07a12aa4241
+size 5304