End of training

README.md

@@ -3,11 +3,23 @@ license: other
 base_model: Qwen/Qwen1.5-4B
 tags:
 - generated_from_trainer
+datasets:
+- tyzhu/lmind_hotpot_train8000_eval7405_v1_qa
 metrics:
 - accuracy
 model-index:
 - name: lmind_hotpot_train8000_eval7405_v1_qa_Qwen_Qwen1.5-4B_lora2
-  results: []
+  results:
+  - task:
+      name: Causal Language Modeling
+      type: text-generation
+    dataset:
+      name: tyzhu/lmind_hotpot_train8000_eval7405_v1_qa
+      type: tyzhu/lmind_hotpot_train8000_eval7405_v1_qa
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.49263492063492065
 library_name: peft
 ---
 
@@ -16,7 +28,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # lmind_hotpot_train8000_eval7405_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_hotpot_train8000_eval7405_v1_qa dataset.
 It achieves the following results on the evaluation set:
 - Loss: 3.4933
 - Accuracy: 0.4926

all_results.json

@@ -0,0 +1,16 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.49263492063492065,
+    "eval_loss": 3.4933342933654785,
+    "eval_runtime": 8.3568,
+    "eval_samples": 500,
+    "eval_samples_per_second": 59.832,
+    "eval_steps_per_second": 7.539,
+    "perplexity": 32.89544812596355,
+    "total_flos": 6.467692909717094e+16,
+    "train_loss": 1.3154203582763673,
+    "train_runtime": 9112.5942,
+    "train_samples": 8000,
+    "train_samples_per_second": 8.779,
+    "train_steps_per_second": 0.274
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.49263492063492065,
+    "eval_loss": 3.4933342933654785,
+    "eval_runtime": 8.3568,
+    "eval_samples": 500,
+    "eval_samples_per_second": 59.832,
+    "eval_steps_per_second": 7.539,
+    "perplexity": 32.89544812596355
+}

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": 10.0,
+    "total_flos": 6.467692909717094e+16,
+    "train_loss": 1.3154203582763673,
+    "train_runtime": 9112.5942,
+    "train_samples": 8000,
+    "train_samples_per_second": 8.779,
+    "train_steps_per_second": 0.274
+}

trainer_state.json

@@ -0,0 +1,295 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 10.0,
+  "eval_steps": 500,
+  "global_step": 2500,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.4,
+      "grad_norm": 0.5425021052360535,
+      "learning_rate": 0.0001,
+      "loss": 2.3548,
+      "step": 100
+    },
+    {
+      "epoch": 0.8,
+      "grad_norm": 0.5427804589271545,
+      "learning_rate": 0.0001,
+      "loss": 2.2624,
+      "step": 200
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.5159047619047619,
+      "eval_loss": 2.3220136165618896,
+      "eval_runtime": 8.2469,
+      "eval_samples_per_second": 60.629,
+      "eval_steps_per_second": 7.639,
+      "step": 250
+    },
+    {
+      "epoch": 1.2,
+      "grad_norm": 0.7659817337989807,
+      "learning_rate": 0.0001,
+      "loss": 2.1894,
+      "step": 300
+    },
+    {
+      "epoch": 1.6,
+      "grad_norm": 0.9417976140975952,
+      "learning_rate": 0.0001,
+      "loss": 2.1155,
+      "step": 400
+    },
+    {
+      "epoch": 2.0,
+      "grad_norm": 0.8788428902626038,
+      "learning_rate": 0.0001,
+      "loss": 2.0942,
+      "step": 500
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.5176190476190476,
+      "eval_loss": 2.32893443107605,
+      "eval_runtime": 6.9616,
+      "eval_samples_per_second": 71.822,
+      "eval_steps_per_second": 9.05,
+      "step": 500
+    },
+    {
+      "epoch": 2.4,
+      "grad_norm": 1.2184635400772095,
+      "learning_rate": 0.0001,
+      "loss": 1.8388,
+      "step": 600
+    },
+    {
+      "epoch": 2.8,
+      "grad_norm": 1.161513090133667,
+      "learning_rate": 0.0001,
+      "loss": 1.8479,
+      "step": 700
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.5147619047619048,
+      "eval_loss": 2.3996665477752686,
+      "eval_runtime": 8.8091,
+      "eval_samples_per_second": 56.76,
+      "eval_steps_per_second": 7.152,
+      "step": 750
+    },
+    {
+      "epoch": 3.2,
+      "grad_norm": 1.4641433954238892,
+      "learning_rate": 0.0001,
+      "loss": 1.7257,
+      "step": 800
+    },
+    {
+      "epoch": 3.6,
+      "grad_norm": 1.4882707595825195,
+      "learning_rate": 0.0001,
+      "loss": 1.5898,
+      "step": 900
+    },
+    {
+      "epoch": 4.0,
+      "grad_norm": 1.650571584701538,
+      "learning_rate": 0.0001,
+      "loss": 1.6153,
+      "step": 1000
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.5106984126984127,
+      "eval_loss": 2.506742000579834,
+      "eval_runtime": 9.1456,
+      "eval_samples_per_second": 54.671,
+      "eval_steps_per_second": 6.889,
+      "step": 1000
+    },
+    {
+      "epoch": 4.4,
+      "grad_norm": 1.7832984924316406,
+      "learning_rate": 0.0001,
+      "loss": 1.328,
+      "step": 1100
+    },
+    {
+      "epoch": 4.8,
+      "grad_norm": 2.0173535346984863,
+      "learning_rate": 0.0001,
+      "loss": 1.3618,
+      "step": 1200
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.5052063492063492,
+      "eval_loss": 2.6640872955322266,
+      "eval_runtime": 7.8899,
+      "eval_samples_per_second": 63.372,
+      "eval_steps_per_second": 7.985,
+      "step": 1250
+    },
+    {
+      "epoch": 5.2,
+      "grad_norm": 1.9236400127410889,
+      "learning_rate": 0.0001,
+      "loss": 1.2386,
+      "step": 1300
+    },
+    {
+      "epoch": 5.6,
+      "grad_norm": 1.9365407228469849,
+      "learning_rate": 0.0001,
+      "loss": 1.1149,
+      "step": 1400
+    },
+    {
+      "epoch": 6.0,
+      "grad_norm": 2.087843656539917,
+      "learning_rate": 0.0001,
+      "loss": 1.1477,
+      "step": 1500
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5015873015873016,
+      "eval_loss": 2.8410720825195312,
+      "eval_runtime": 7.5175,
+      "eval_samples_per_second": 66.512,
+      "eval_steps_per_second": 8.38,
+      "step": 1500
+    },
+    {
+      "epoch": 6.4,
+      "grad_norm": 2.397096872329712,
+      "learning_rate": 0.0001,
+      "loss": 0.8925,
+      "step": 1600
+    },
+    {
+      "epoch": 6.8,
+      "grad_norm": 2.441624641418457,
+      "learning_rate": 0.0001,
+      "loss": 0.9248,
+      "step": 1700
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.49777777777777776,
+      "eval_loss": 3.0245563983917236,
+      "eval_runtime": 8.2076,
+      "eval_samples_per_second": 60.919,
+      "eval_steps_per_second": 7.676,
+      "step": 1750
+    },
+    {
+      "epoch": 7.2,
+      "grad_norm": 2.44694185256958,
+      "learning_rate": 0.0001,
+      "loss": 0.821,
+      "step": 1800
+    },
+    {
+      "epoch": 7.6,
+      "grad_norm": 2.26471209526062,
+      "learning_rate": 0.0001,
+      "loss": 0.7379,
+      "step": 1900
+    },
+    {
+      "epoch": 8.0,
+      "grad_norm": 2.3444228172302246,
+      "learning_rate": 0.0001,
+      "loss": 0.7705,
+      "step": 2000
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.4953968253968254,
+      "eval_loss": 3.2090303897857666,
+      "eval_runtime": 7.2355,
+      "eval_samples_per_second": 69.104,
+      "eval_steps_per_second": 8.707,
+      "step": 2000
+    },
+    {
+      "epoch": 8.4,
+      "grad_norm": 2.152717113494873,
+      "learning_rate": 0.0001,
+      "loss": 0.601,
+      "step": 2100
+    },
+    {
+      "epoch": 8.8,
+      "grad_norm": 2.0435783863067627,
+      "learning_rate": 0.0001,
+      "loss": 0.6344,
+      "step": 2200
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.4934920634920635,
+      "eval_loss": 3.339984893798828,
+      "eval_runtime": 7.5883,
+      "eval_samples_per_second": 65.891,
+      "eval_steps_per_second": 8.302,
+      "step": 2250
+    },
+    {
+      "epoch": 9.2,
+      "grad_norm": 1.7771090269088745,
+      "learning_rate": 0.0001,
+      "loss": 0.5798,
+      "step": 2300
+    },
+    {
+      "epoch": 9.6,
+      "grad_norm": 1.9982362985610962,
+      "learning_rate": 0.0001,
+      "loss": 0.5377,
+      "step": 2400
+    },
+    {
+      "epoch": 10.0,
+      "grad_norm": 1.932521104812622,
+      "learning_rate": 0.0001,
+      "loss": 0.5612,
+      "step": 2500
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.49263492063492065,
+      "eval_loss": 3.4933342933654785,
+      "eval_runtime": 7.1358,
+      "eval_samples_per_second": 70.069,
+      "eval_steps_per_second": 8.829,
+      "step": 2500
+    },
+    {
+      "epoch": 10.0,
+      "step": 2500,
+      "total_flos": 6.467692909717094e+16,
+      "train_loss": 1.3154203582763673,
+      "train_runtime": 9112.5942,
+      "train_samples_per_second": 8.779,
+      "train_steps_per_second": 0.274
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 2500,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "total_flos": 6.467692909717094e+16,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}