End of training

README.md

@@ -3,11 +3,23 @@ license: other
 base_model: Qwen/Qwen1.5-4B
 tags:
 - generated_from_trainer
+datasets:
+- tyzhu/lmind_nq_train6000_eval6489_v1_qa
 metrics:
 - accuracy
 model-index:
 - name: lmind_nq_train6000_eval6489_v1_qa_Qwen_Qwen1.5-4B_lora2
-  results: []
+  results:
+  - task:
+      name: Causal Language Modeling
+      type: text-generation
+    dataset:
+      name: tyzhu/lmind_nq_train6000_eval6489_v1_qa
+      type: tyzhu/lmind_nq_train6000_eval6489_v1_qa
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.5594358974358974
 library_name: peft
 ---
 
@@ -16,7 +28,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # lmind_nq_train6000_eval6489_v1_qa_Qwen_Qwen1.5-4B_lora2
 
-This model is a fine-tuned version of [Qwen/Qwen1.5-4B](https://huggingface.co/Qwen/Qwen1.5-4B) on an unknown dataset.
+This model is a fine-tuned version of [Qwen/Qwen1.5-4B](https://huggingface.co/Qwen/Qwen1.5-4B) on the tyzhu/lmind_nq_train6000_eval6489_v1_qa dataset.
 It achieves the following results on the evaluation set:
 - Loss: 2.2527
 - Accuracy: 0.5594

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 9.973333333333333,
+    "eval_accuracy": 0.5594358974358974,
+    "eval_loss": 2.252700090408325,
+    "eval_runtime": 7.2866,
+    "eval_samples": 500,
+    "eval_samples_per_second": 68.619,
+    "eval_steps_per_second": 8.646,
+    "perplexity": 9.51338819714881,
+    "total_flos": 3.256692486491341e+16,
+    "train_loss": 0.933582401785621,
+    "train_runtime": 6553.864,
+    "train_samples": 6000,
+    "train_samples_per_second": 9.155,
+    "train_steps_per_second": 0.285
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 9.973333333333333,
+    "eval_accuracy": 0.5594358974358974,
+    "eval_loss": 2.252700090408325,
+    "eval_runtime": 7.2866,
+    "eval_samples": 500,
+    "eval_samples_per_second": 68.619,
+    "eval_steps_per_second": 8.646,
+    "perplexity": 9.51338819714881
+}

tokenizer.json

@@ -1,11 +1,6 @@
 {
   "version": "1.0",
-  "truncation": {
-    "direction": "Right",
-    "max_length": 1024,
-    "strategy": "LongestFirst",
-    "stride": 0
-  },
+  "truncation": null,
   "padding": null,
   "added_tokens": [
     {

train_results.json

@@ -0,0 +1,9 @@
+{
+    "epoch": 9.973333333333333,
+    "total_flos": 3.256692486491341e+16,
+    "train_loss": 0.933582401785621,
+    "train_runtime": 6553.864,
+    "train_samples": 6000,
+    "train_samples_per_second": 9.155,
+    "train_steps_per_second": 0.285
+}

trainer_state.json

@@ -0,0 +1,246 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 9.973333333333333,
+  "eval_steps": 500,
+  "global_step": 1870,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.5333333333333333,
+      "grad_norm": 0.5919559597969055,
+      "learning_rate": 0.0001,
+      "loss": 1.7657,
+      "step": 100
+    },
+    {
+      "epoch": 0.9973333333333333,
+      "eval_accuracy": 0.5738461538461539,
+      "eval_loss": 1.6214678287506104,
+      "eval_runtime": 6.7752,
+      "eval_samples_per_second": 73.799,
+      "eval_steps_per_second": 9.299,
+      "step": 187
+    },
+    {
+      "epoch": 1.0666666666666667,
+      "grad_norm": 0.7285414934158325,
+      "learning_rate": 0.0001,
+      "loss": 1.6421,
+      "step": 200
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 1.0658913850784302,
+      "learning_rate": 0.0001,
+      "loss": 1.497,
+      "step": 300
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5742051282051283,
+      "eval_loss": 1.6179509162902832,
+      "eval_runtime": 7.0059,
+      "eval_samples_per_second": 71.369,
+      "eval_steps_per_second": 8.992,
+      "step": 375
+    },
+    {
+      "epoch": 2.1333333333333333,
+      "grad_norm": 1.2808786630630493,
+      "learning_rate": 0.0001,
+      "loss": 1.4367,
+      "step": 400
+    },
+    {
+      "epoch": 2.6666666666666665,
+      "grad_norm": 1.369543194770813,
+      "learning_rate": 0.0001,
+      "loss": 1.2345,
+      "step": 500
+    },
+    {
+      "epoch": 2.997333333333333,
+      "eval_accuracy": 0.5712820512820512,
+      "eval_loss": 1.6951237916946411,
+      "eval_runtime": 7.4766,
+      "eval_samples_per_second": 66.875,
+      "eval_steps_per_second": 8.426,
+      "step": 562
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 1.4907397031784058,
+      "learning_rate": 0.0001,
+      "loss": 1.157,
+      "step": 600
+    },
+    {
+      "epoch": 3.7333333333333334,
+      "grad_norm": 1.5697628259658813,
+      "learning_rate": 0.0001,
+      "loss": 1.0084,
+      "step": 700
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5658974358974359,
+      "eval_loss": 1.8059020042419434,
+      "eval_runtime": 7.0878,
+      "eval_samples_per_second": 70.544,
+      "eval_steps_per_second": 8.889,
+      "step": 750
+    },
+    {
+      "epoch": 4.266666666666667,
+      "grad_norm": 1.6422948837280273,
+      "learning_rate": 0.0001,
+      "loss": 0.9221,
+      "step": 800
+    },
+    {
+      "epoch": 4.8,
+      "grad_norm": 1.6770026683807373,
+      "learning_rate": 0.0001,
+      "loss": 0.8397,
+      "step": 900
+    },
+    {
+      "epoch": 4.997333333333334,
+      "eval_accuracy": 0.5646666666666667,
+      "eval_loss": 1.9245407581329346,
+      "eval_runtime": 7.4653,
+      "eval_samples_per_second": 66.977,
+      "eval_steps_per_second": 8.439,
+      "step": 937
+    },
+    {
+      "epoch": 5.333333333333333,
+      "grad_norm": 1.588730812072754,
+      "learning_rate": 0.0001,
+      "loss": 0.7486,
+      "step": 1000
+    },
+    {
+      "epoch": 5.866666666666667,
+      "grad_norm": 1.8069469928741455,
+      "learning_rate": 0.0001,
+      "loss": 0.7186,
+      "step": 1100
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5614358974358974,
+      "eval_loss": 2.0345287322998047,
+      "eval_runtime": 6.285,
+      "eval_samples_per_second": 79.554,
+      "eval_steps_per_second": 10.024,
+      "step": 1125
+    },
+    {
+      "epoch": 6.4,
+      "grad_norm": 1.590628981590271,
+      "learning_rate": 0.0001,
+      "loss": 0.6356,
+      "step": 1200
+    },
+    {
+      "epoch": 6.933333333333334,
+      "grad_norm": 1.9079328775405884,
+      "learning_rate": 0.0001,
+      "loss": 0.6421,
+      "step": 1300
+    },
+    {
+      "epoch": 6.997333333333334,
+      "eval_accuracy": 0.5607692307692308,
+      "eval_loss": 2.1147842407226562,
+      "eval_runtime": 8.0598,
+      "eval_samples_per_second": 62.037,
+      "eval_steps_per_second": 7.817,
+      "step": 1312
+    },
+    {
+      "epoch": 7.466666666666667,
+      "grad_norm": 1.3450945615768433,
+      "learning_rate": 0.0001,
+      "loss": 0.5814,
+      "step": 1400
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 1.2447816133499146,
+      "learning_rate": 0.0001,
+      "loss": 0.5968,
+      "step": 1500
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5584615384615385,
+      "eval_loss": 2.1779356002807617,
+      "eval_runtime": 7.4067,
+      "eval_samples_per_second": 67.506,
+      "eval_steps_per_second": 8.506,
+      "step": 1500
+    },
+    {
+      "epoch": 8.533333333333333,
+      "grad_norm": 1.113315463066101,
+      "learning_rate": 0.0001,
+      "loss": 0.5417,
+      "step": 1600
+    },
+    {
+      "epoch": 8.997333333333334,
+      "eval_accuracy": 0.5567692307692308,
+      "eval_loss": 2.265408754348755,
+      "eval_runtime": 6.4222,
+      "eval_samples_per_second": 77.854,
+      "eval_steps_per_second": 9.81,
+      "step": 1687
+    },
+    {
+      "epoch": 9.066666666666666,
+      "grad_norm": 2.0787792205810547,
+      "learning_rate": 0.0001,
+      "loss": 0.5678,
+      "step": 1700
+    },
+    {
+      "epoch": 9.6,
+      "grad_norm": 1.027565598487854,
+      "learning_rate": 0.0001,
+      "loss": 0.5356,
+      "step": 1800
+    },
+    {
+      "epoch": 9.973333333333333,
+      "eval_accuracy": 0.5594358974358974,
+      "eval_loss": 2.252700090408325,
+      "eval_runtime": 7.737,
+      "eval_samples_per_second": 64.624,
+      "eval_steps_per_second": 8.143,
+      "step": 1870
+    },
+    {
+      "epoch": 9.973333333333333,
+      "step": 1870,
+      "total_flos": 3.256692486491341e+16,
+      "train_loss": 0.933582401785621,
+      "train_runtime": 6553.864,
+      "train_samples_per_second": 9.155,
+      "train_steps_per_second": 0.285
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 1870,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "total_flos": 3.256692486491341e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}