End of training

README.md

@@ -1,17 +1,21 @@
 ---
 base_model: HuggingFaceTB/SmolLM2-135M-Instruct
+datasets: Crimsoin/OTC_Medicine_PH_v2
 library_name: transformers
-model_name: HFTB-SmolLM2-135M-Instruct-OTCMedicinePHv2
+model_name: HuggingFaceTB/SmolLM2-135M-Instruct
 tags:
 - generated_from_trainer
+- question-answering
+- QA
+- text-generation
 - trl
 - sft
 licence: license
 ---
 
-# Model Card for HFTB-SmolLM2-135M-Instruct-OTCMedicinePHv2
+# Model Card for HuggingFaceTB/SmolLM2-135M-Instruct
 
-This model is a fine-tuned version of [HuggingFaceTB/SmolLM2-135M-Instruct](https://huggingface.co/HuggingFaceTB/SmolLM2-135M-Instruct).
+This model is a fine-tuned version of [HuggingFaceTB/SmolLM2-135M-Instruct](https://huggingface.co/HuggingFaceTB/SmolLM2-135M-Instruct) on the [Crimsoin/OTC_Medicine_PH_v2](https://huggingface.co/datasets/Crimsoin/OTC_Medicine_PH_v2) dataset.
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start

all_results.json

@@ -1,12 +1,12 @@
 {
     "epoch": 3.883495145631068,
-    "eval_loss": 0.1291082799434662,
-    "eval_runtime": 14.8078,
-    "eval_samples_per_second": 35.792,
-    "eval_steps_per_second": 4.525,
-    "total_flos": 258445456835328.0,
-    "train_loss": 0.47970603942871093,
-    "train_runtime": 930.7066,
-    "train_samples_per_second": 5.157,
-    "train_steps_per_second": 0.107
+    "eval_loss": 0.14737741649150848,
+    "eval_runtime": 14.0975,
+    "eval_samples_per_second": 37.595,
+    "eval_steps_per_second": 4.753,
+    "total_flos": 258162252195456.0,
+    "train_loss": 0.4854153060913086,
+    "train_runtime": 885.215,
+    "train_samples_per_second": 5.422,
+    "train_steps_per_second": 0.113
 }

eval_results.json

@@ -1,7 +1,7 @@
 {
     "epoch": 3.883495145631068,
-    "eval_loss": 0.1291082799434662,
-    "eval_runtime": 14.8078,
-    "eval_samples_per_second": 35.792,
-    "eval_steps_per_second": 4.525
+    "eval_loss": 0.14737741649150848,
+    "eval_runtime": 14.0975,
+    "eval_samples_per_second": 37.595,
+    "eval_steps_per_second": 4.753
 }

train_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 3.883495145631068,
-    "total_flos": 258445456835328.0,
-    "train_loss": 0.47970603942871093,
-    "train_runtime": 930.7066,
-    "train_samples_per_second": 5.157,
-    "train_steps_per_second": 0.107
+    "total_flos": 258162252195456.0,
+    "train_loss": 0.4854153060913086,
+    "train_runtime": 885.215,
+    "train_samples_per_second": 5.422,
+    "train_steps_per_second": 0.113
 }

trainer_state.json

@@ -1,5 +1,5 @@
 {
-  "best_metric": 0.12910962104797363,
+  "best_metric": 0.14742426574230194,
   "best_model_checkpoint": "./HFTB-SmolLM2-135M-Instruct-OTCMedicinePHv2\\checkpoint-100",
   "epoch": 3.883495145631068,
   "eval_steps": 100,
@@ -10,27 +10,27 @@
   "log_history": [
     {
       "epoch": 3.883495145631068,
-      "grad_norm": 0.26728370785713196,
+      "grad_norm": 0.20863692462444305,
       "learning_rate": 0.0005,
-      "loss": 0.4797,
+      "loss": 0.4854,
       "step": 100
     },
     {
       "epoch": 3.883495145631068,
-      "eval_loss": 0.12910962104797363,
-      "eval_runtime": 14.3923,
-      "eval_samples_per_second": 36.825,
-      "eval_steps_per_second": 4.655,
+      "eval_loss": 0.14742426574230194,
+      "eval_runtime": 13.1852,
+      "eval_samples_per_second": 40.197,
+      "eval_steps_per_second": 5.081,
       "step": 100
     },
     {
       "epoch": 3.883495145631068,
       "step": 100,
-      "total_flos": 258445456835328.0,
-      "train_loss": 0.47970603942871093,
-      "train_runtime": 930.7066,
-      "train_samples_per_second": 5.157,
-      "train_steps_per_second": 0.107
+      "total_flos": 258162252195456.0,
+      "train_loss": 0.4854153060913086,
+      "train_runtime": 885.215,
+      "train_samples_per_second": 5.422,
+      "train_steps_per_second": 0.113
     }
   ],
   "logging_steps": 100,
@@ -59,7 +59,7 @@
       "attributes": {}
     }
   },
-  "total_flos": 258445456835328.0,
+  "total_flos": 258162252195456.0,
   "train_batch_size": 4,
   "trial_name": null,
   "trial_params": null