End of training

README.md

@@ -2,11 +2,15 @@
 license: mit
 base_model: microsoft/phi-2
 tags:
+- alignment-handbook
+- trl
+- sft
+- generated_from_trainer
 - trl
 - sft
 - generated_from_trainer
 datasets:
-- generator
+- GAIR/lima
 model-index:
 - name: phi2-lima
   results: []
@@ -17,7 +21,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # phi2-lima
 
-This model is a fine-tuned version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) on the generator dataset.
+This model is a fine-tuned version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) on the GAIR/lima dataset.
 It achieves the following results on the evaluation set:
 - Loss: 2.5096
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 10.0,
+    "eval_loss": 2.5095889568328857,
+    "eval_runtime": 0.8853,
+    "eval_samples": 300,
+    "eval_samples_per_second": 11.295,
+    "eval_steps_per_second": 1.13,
     "train_loss": 1.8223904927571615,
     "train_runtime": 196.3016,
     "train_samples": 1030,

config.json

@@ -29,6 +29,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.38.2",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 51200
 }

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 10.0,
+    "eval_loss": 2.5095889568328857,
+    "eval_runtime": 0.8853,
+    "eval_samples": 300,
+    "eval_samples_per_second": 11.295,
+    "eval_steps_per_second": 1.13
+}

runs/Apr26_16-44-19_aga43/events.out.tfevents.1714168262.aga43.1097587.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:abaf4441b0b82ebf924e0fb19fd94e02fdcf18ff5290a2ada169531c3c9a54ab
+size 354