End of training

README.md

@@ -3,9 +3,15 @@ library_name: transformers
 license: apache-2.0
 base_model: EleutherAI/pythia-70m
 tags:
+- alignment-handbook
 - trl
 - sft
 - generated_from_trainer
+- trl
+- sft
+- generated_from_trainer
+datasets:
+- allenai/tulu-v2-sft-mixture
 model-index:
 - name: pythia-70m-tulu-v2-mix
   results: []
@@ -16,7 +22,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # pythia-70m-tulu-v2-mix
 
-This model is a fine-tuned version of [EleutherAI/pythia-70m](https://huggingface.co/EleutherAI/pythia-70m) on the None dataset.
+This model is a fine-tuned version of [EleutherAI/pythia-70m](https://huggingface.co/EleutherAI/pythia-70m) on the allenai/tulu-v2-sft-mixture dataset.
 It achieves the following results on the evaluation set:
 - Loss: 3.0855
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 2.999705680368881,
+    "eval_loss": 3.0855159759521484,
+    "eval_runtime": 89.1192,
+    "eval_samples": 25000,
+    "eval_samples_per_second": 280.523,
+    "eval_steps_per_second": 70.131,
     "total_flos": 2.04275798866133e+17,
     "train_loss": 3.3574186354516877,
     "train_runtime": 4757.4793,

config.json

@@ -25,7 +25,7 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.44.2",
-  "use_cache": false,
+  "use_cache": true,
   "use_parallel_residual": true,
   "vocab_size": 50304
 }

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.999705680368881,
+    "eval_loss": 3.0855159759521484,
+    "eval_runtime": 89.1192,
+    "eval_samples": 25000,
+    "eval_samples_per_second": 280.523,
+    "eval_steps_per_second": 70.131
+}