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_nq_train6000_eval6489_v1_docidx_v3
 metrics:
 - accuracy
 model-index:
 - name: lmind_nq_train6000_eval6489_v1_docidx_v3_5e-4_lora2
-  results: []
+  results:
+  - task:
+      name: Causal Language Modeling
+      type: text-generation
+    dataset:
+      name: tyzhu/lmind_nq_train6000_eval6489_v1_docidx_v3
+      type: tyzhu/lmind_nq_train6000_eval6489_v1_docidx_v3
+    metrics:
+    - name: Accuracy
+      type: accuracy
+      value: 0.19446153846153846
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -15,10 +27,10 @@ should probably proofread and complete it, then remove this comment. -->
 
 # lmind_nq_train6000_eval6489_v1_docidx_v3_5e-4_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_nq_train6000_eval6489_v1_docidx_v3 dataset.
 It achieves the following results on the evaluation set:
-- Accuracy: 0.1945
 - Loss: 8.0353
+- Accuracy: 0.1945
 
 ## Model description
 

all_results.json

@@ -1,8 +1,18 @@
 {
     "epoch": 49.93,
+    "eval_accuracy": 0.19446153846153846,
+    "eval_bleu": 0.0025420935228995492,
+    "eval_exact_match": 0.0,
+    "eval_loss": 8.035279273986816,
+    "eval_prefix_exact_match": 0.0,
+    "eval_runtime": 4.5946,
+    "eval_samples": 500,
+    "eval_samples_per_second": 108.824,
+    "eval_steps_per_second": 13.712,
+    "perplexity": 3088.0011204567613,
     "train_loss": 0.0,
-    "train_runtime": 0.0625,
+    "train_runtime": 0.0599,
     "train_samples": 10925,
-    "train_samples_per_second": 8739133.234,
-    "train_steps_per_second": 272772.946
+    "train_samples_per_second": 9121537.071,
+    "train_steps_per_second": 284708.846
 }

eval_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 49.926793557833086,
-    "eval_accuracy": 0.38,
-    "eval_bleu": 0.20225217833889003,
+    "epoch": 49.93,
+    "eval_accuracy": 0.19446153846153846,
+    "eval_bleu": 0.0025420935228995492,
     "eval_exact_match": 0.0,
-    "eval_loss": 6.628392219543457,
-    "eval_prefix_exact_match": 0.024,
-    "eval_runtime": 5.1985,
+    "eval_loss": 8.035279273986816,
+    "eval_prefix_exact_match": 0.0,
+    "eval_runtime": 4.5946,
     "eval_samples": 500,
-    "eval_samples_per_second": 96.182,
-    "eval_steps_per_second": 12.119,
-    "perplexity": 756.2652841169914
+    "eval_samples_per_second": 108.824,
+    "eval_steps_per_second": 13.712,
+    "perplexity": 3088.0011204567613
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 49.93,
     "train_loss": 0.0,
-    "train_runtime": 0.0625,
+    "train_runtime": 0.0599,
     "train_samples": 10925,
-    "train_samples_per_second": 8739133.234,
-    "train_steps_per_second": 272772.946
+    "train_samples_per_second": 9121537.071,
+    "train_steps_per_second": 284708.846
 }

trainer_state.json

@@ -1830,9 +1830,9 @@
       "step": 17050,
       "total_flos": 3.3761617516943114e+18,
       "train_loss": 0.0,
-      "train_runtime": 0.0625,
-      "train_samples_per_second": 8739133.234,
-      "train_steps_per_second": 272772.946
+      "train_runtime": 0.0599,
+      "train_samples_per_second": 9121537.071,
+      "train_steps_per_second": 284708.846
     }
   ],
   "logging_steps": 100,