End of training

README.md

@@ -3,11 +3,23 @@ license: llama2
 base_model: meta-llama/Llama-2-7b-hf
 tags:
 - generated_from_trainer
+datasets:
+- tyzhu/lmind_hotpot_train8000_eval7405_v1_qa
 metrics:
 - accuracy
 model-index:
 - name: lmind_hotpot_train8000_eval7405_v1_qa_meta-llama_Llama-2-7b-hf_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: 0.5903037974683545
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -15,7 +27,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # lmind_hotpot_train8000_eval7405_v1_qa_meta-llama_Llama-2-7b-hf_lora2
 
-This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on an unknown dataset.
+This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on the tyzhu/lmind_hotpot_train8000_eval7405_v1_qa dataset.
 It achieves the following results on the evaluation set:
 - Loss: 2.8216
 - Accuracy: 0.5903

all_results.json

@@ -0,0 +1,15 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.5903037974683545,
+    "eval_loss": 2.8216025829315186,
+    "eval_runtime": 4.5487,
+    "eval_samples": 500,
+    "eval_samples_per_second": 109.921,
+    "eval_steps_per_second": 13.85,
+    "perplexity": 16.8037585219432,
+    "train_loss": 0.8466211395263672,
+    "train_runtime": 1749.1622,
+    "train_samples": 8000,
+    "train_samples_per_second": 45.736,
+    "train_steps_per_second": 1.429
+}

eval_results.json

@@ -0,0 +1,10 @@
+{
+    "epoch": 10.0,
+    "eval_accuracy": 0.5903037974683545,
+    "eval_loss": 2.8216025829315186,
+    "eval_runtime": 4.5487,
+    "eval_samples": 500,
+    "eval_samples_per_second": 109.921,
+    "eval_steps_per_second": 13.85,
+    "perplexity": 16.8037585219432
+}

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 +1,8 @@
+{
+    "epoch": 10.0,
+    "train_loss": 0.8466211395263672,
+    "train_runtime": 1749.1622,
+    "train_samples": 8000,
+    "train_samples_per_second": 45.736,
+    "train_steps_per_second": 1.429
+}

trainer_state.json

@@ -0,0 +1,274 @@
+{
+  "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,
+      "learning_rate": 0.0001,
+      "loss": 2.0275,
+      "step": 100
+    },
+    {
+      "epoch": 0.8,
+      "learning_rate": 0.0001,
+      "loss": 1.7795,
+      "step": 200
+    },
+    {
+      "epoch": 1.0,
+      "eval_accuracy": 0.6075443037974684,
+      "eval_loss": 1.8062162399291992,
+      "eval_runtime": 23.0836,
+      "eval_samples_per_second": 21.66,
+      "eval_steps_per_second": 2.729,
+      "step": 250
+    },
+    {
+      "epoch": 1.0,
+      "eval_exact_match": 23.686698176907495,
+      "eval_f1": 33.6125461509162,
+      "step": 250
+    },
+    {
+      "epoch": 1.2,
+      "learning_rate": 0.0001,
+      "loss": 1.66,
+      "step": 300
+    },
+    {
+      "epoch": 1.6,
+      "learning_rate": 0.0001,
+      "loss": 1.6532,
+      "step": 400
+    },
+    {
+      "epoch": 2.0,
+      "learning_rate": 0.0001,
+      "loss": 1.6437,
+      "step": 500
+    },
+    {
+      "epoch": 2.0,
+      "eval_accuracy": 0.6077215189873417,
+      "eval_loss": 1.8113620281219482,
+      "eval_runtime": 4.6236,
+      "eval_samples_per_second": 108.141,
+      "eval_steps_per_second": 13.626,
+      "step": 500
+    },
+    {
+      "epoch": 2.4,
+      "learning_rate": 0.0001,
+      "loss": 1.438,
+      "step": 600
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 0.0001,
+      "loss": 1.4652,
+      "step": 700
+    },
+    {
+      "epoch": 3.0,
+      "eval_accuracy": 0.6060759493670886,
+      "eval_loss": 1.8674805164337158,
+      "eval_runtime": 4.715,
+      "eval_samples_per_second": 106.043,
+      "eval_steps_per_second": 13.361,
+      "step": 750
+    },
+    {
+      "epoch": 3.2,
+      "learning_rate": 0.0001,
+      "loss": 1.3476,
+      "step": 800
+    },
+    {
+      "epoch": 3.6,
+      "learning_rate": 0.0001,
+      "loss": 1.2363,
+      "step": 900
+    },
+    {
+      "epoch": 4.0,
+      "learning_rate": 0.0001,
+      "loss": 1.2631,
+      "step": 1000
+    },
+    {
+      "epoch": 4.0,
+      "eval_accuracy": 0.6029873417721519,
+      "eval_loss": 1.9842653274536133,
+      "eval_runtime": 4.5617,
+      "eval_samples_per_second": 109.607,
+      "eval_steps_per_second": 13.811,
+      "step": 1000
+    },
+    {
+      "epoch": 4.4,
+      "learning_rate": 0.0001,
+      "loss": 1.0182,
+      "step": 1100
+    },
+    {
+      "epoch": 4.8,
+      "learning_rate": 0.0001,
+      "loss": 1.0724,
+      "step": 1200
+    },
+    {
+      "epoch": 5.0,
+      "eval_accuracy": 0.600126582278481,
+      "eval_loss": 2.092109441757202,
+      "eval_runtime": 4.6632,
+      "eval_samples_per_second": 107.223,
+      "eval_steps_per_second": 13.51,
+      "step": 1250
+    },
+    {
+      "epoch": 5.2,
+      "learning_rate": 0.0001,
+      "loss": 0.9575,
+      "step": 1300
+    },
+    {
+      "epoch": 5.6,
+      "learning_rate": 0.0001,
+      "loss": 0.8617,
+      "step": 1400
+    },
+    {
+      "epoch": 6.0,
+      "learning_rate": 0.0001,
+      "loss": 0.8917,
+      "step": 1500
+    },
+    {
+      "epoch": 6.0,
+      "eval_accuracy": 0.5973164556962025,
+      "eval_loss": 2.2463245391845703,
+      "eval_runtime": 4.7168,
+      "eval_samples_per_second": 106.004,
+      "eval_steps_per_second": 13.357,
+      "step": 1500
+    },
+    {
+      "epoch": 6.4,
+      "learning_rate": 0.0001,
+      "loss": 0.6818,
+      "step": 1600
+    },
+    {
+      "epoch": 6.8,
+      "learning_rate": 0.0001,
+      "loss": 0.7235,
+      "step": 1700
+    },
+    {
+      "epoch": 7.0,
+      "eval_accuracy": 0.5943291139240506,
+      "eval_loss": 2.4073448181152344,
+      "eval_runtime": 4.8002,
+      "eval_samples_per_second": 104.162,
+      "eval_steps_per_second": 13.124,
+      "step": 1750
+    },
+    {
+      "epoch": 7.2,
+      "learning_rate": 0.0001,
+      "loss": 0.6412,
+      "step": 1800
+    },
+    {
+      "epoch": 7.6,
+      "learning_rate": 0.0001,
+      "loss": 0.575,
+      "step": 1900
+    },
+    {
+      "epoch": 8.0,
+      "learning_rate": 0.0001,
+      "loss": 0.5997,
+      "step": 2000
+    },
+    {
+      "epoch": 8.0,
+      "eval_accuracy": 0.5931392405063292,
+      "eval_loss": 2.573843240737915,
+      "eval_runtime": 5.5843,
+      "eval_samples_per_second": 89.537,
+      "eval_steps_per_second": 11.282,
+      "step": 2000
+    },
+    {
+      "epoch": 8.4,
+      "learning_rate": 0.0001,
+      "loss": 0.4636,
+      "step": 2100
+    },
+    {
+      "epoch": 8.8,
+      "learning_rate": 0.0001,
+      "loss": 0.4943,
+      "step": 2200
+    },
+    {
+      "epoch": 9.0,
+      "eval_accuracy": 0.5905316455696202,
+      "eval_loss": 2.6983227729797363,
+      "eval_runtime": 4.93,
+      "eval_samples_per_second": 101.42,
+      "eval_steps_per_second": 12.779,
+      "step": 2250
+    },
+    {
+      "epoch": 9.2,
+      "learning_rate": 0.0001,
+      "loss": 0.4472,
+      "step": 2300
+    },
+    {
+      "epoch": 9.6,
+      "learning_rate": 0.0001,
+      "loss": 0.4226,
+      "step": 2400
+    },
+    {
+      "epoch": 10.0,
+      "learning_rate": 0.0001,
+      "loss": 0.4381,
+      "step": 2500
+    },
+    {
+      "epoch": 10.0,
+      "eval_accuracy": 0.5903037974683545,
+      "eval_loss": 2.8216025829315186,
+      "eval_runtime": 4.647,
+      "eval_samples_per_second": 107.597,
+      "eval_steps_per_second": 13.557,
+      "step": 2500
+    },
+    {
+      "epoch": 10.0,
+      "step": 2500,
+      "total_flos": 1.6603500238497382e+17,
+      "train_loss": 0.8466211395263672,
+      "train_runtime": 1749.1622,
+      "train_samples_per_second": 45.736,
+      "train_steps_per_second": 1.429
+    }
+  ],
+  "logging_steps": 100,
+  "max_steps": 2500,
+  "num_train_epochs": 10,
+  "save_steps": 500,
+  "total_flos": 1.6603500238497382e+17,
+  "trial_name": null,
+  "trial_params": null
+}