End of training

README.md

@@ -3,11 +3,15 @@ library_name: transformers
 license: other
 base_model: apple/OpenELM-1_1B
 tags:
+- alignment-handbook
+- trl
+- sft
+- generated_from_trainer
 - trl
 - sft
 - generated_from_trainer
 datasets:
-- generator
+- HuggingFaceH4/ultrachat_200k
 model-index:
 - name: OpenELM-1_1B-SFT
   results: []
@@ -18,7 +22,9 @@ should probably proofread and complete it, then remove this comment. -->
 
 # OpenELM-1_1B-SFT
 
-This model is a fine-tuned version of [apple/OpenELM-1_1B](https://huggingface.co/apple/OpenELM-1_1B) on the generator dataset.
+This model is a fine-tuned version of [apple/OpenELM-1_1B](https://huggingface.co/apple/OpenELM-1_1B) on the HuggingFaceH4/ultrachat_200k dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.1288
 
 ## Model description
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 0.9995619798510732,
+    "eval_loss": 1.1288038492202759,
+    "eval_runtime": 220.1694,
+    "eval_samples": 23109,
+    "eval_samples_per_second": 73.425,
+    "eval_steps_per_second": 2.298,
     "total_flos": 215836199485440.0,
     "train_loss": 1.3199829850460123,
     "train_runtime": 8951.0623,

config.json

@@ -120,6 +120,6 @@
   "share_input_output_layers": true,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.44.2",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 32000
 }

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 0.9995619798510732,
+    "eval_loss": 1.1288038492202759,
+    "eval_runtime": 220.1694,
+    "eval_samples": 23109,
+    "eval_samples_per_second": 73.425,
+    "eval_steps_per_second": 2.298
+}

runs/Aug27_08-29-06_xe8545-a100-16/events.out.tfevents.1724750476.xe8545-a100-16.2577838.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2b7457e95d3d3fecbe7ca000b0dde45b009109bdcb4585343dd55b6b6881b4ed
+size 359