End of training

README.md

@@ -2,12 +2,16 @@
 license: apache-2.0
 base_model: mistralai/Mistral-7B-v0.1
 tags:
+- alignment-handbook
+- trl
+- sft
+- generated_from_trainer
 - trl
 - sft
 - alignment-handbook
 - generated_from_trainer
 datasets:
-- generator
+- AmberYifan/safeRLHF-SFT
 model-index:
 - name: mistral-safe-sft-full
   results: []
@@ -18,7 +22,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # mistral-safe-sft-full
 
-This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the generator dataset.
+This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on the AmberYifan/safeRLHF-SFT dataset.
 It achieves the following results on the evaluation set:
 - Loss: 1.7972
 

all_results.json

@@ -1,10 +1,10 @@
 {
     "epoch": 4.0,
-    "eval_loss": 1.2626361846923828,
-    "eval_runtime": 50.2072,
+    "eval_loss": 1.7972235679626465,
+    "eval_runtime": 52.7167,
     "eval_samples": 8340,
-    "eval_samples_per_second": 13.126,
-    "eval_steps_per_second": 0.418,
+    "eval_samples_per_second": 12.501,
+    "eval_steps_per_second": 0.398,
     "total_flos": 77470472601600.0,
     "train_loss": 0.38161755220310106,
     "train_runtime": 4944.5179,

config.json

@@ -22,6 +22,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.43.3",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 32000
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 1.0,
-    "eval_loss": 1.2626361846923828,
-    "eval_runtime": 50.2072,
+    "epoch": 4.0,
+    "eval_loss": 1.7972235679626465,
+    "eval_runtime": 52.7167,
     "eval_samples": 8340,
-    "eval_samples_per_second": 13.126,
-    "eval_steps_per_second": 0.418
+    "eval_samples_per_second": 12.501,
+    "eval_steps_per_second": 0.398
 }