End of training

README.md

@@ -2,11 +2,16 @@
 license: llama2
 base_model: meta-llama/Llama-2-7b-hf
 tags:
+- alignment-handbook
+- trl
+- sft
+- generated_from_trainer
 - trl
 - sft
 - generated_from_trainer
 datasets:
-- generator
+- HuggingFaceH4/ultrachat_200k
+- HuggingFaceH4/ultrafeedback_binarized
 model-index:
 - name: llama2-7b-sft-full
   results: []
@@ -17,7 +22,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # llama2-7b-sft-full
 
-This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on the generator dataset.
+This model is a fine-tuned version of [meta-llama/Llama-2-7b-hf](https://huggingface.co/meta-llama/Llama-2-7b-hf) on the HuggingFaceH4/ultrachat_200k and the HuggingFaceH4/ultrafeedback_binarized datasets.
 It achieves the following results on the evaluation set:
 - Loss: 0.9454
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 2.9988175009854157,
+    "eval_loss": 0.9453620314598083,
+    "eval_runtime": 625.2253,
+    "eval_samples": 24109,
+    "eval_samples_per_second": 26.285,
+    "eval_steps_per_second": 1.644,
     "total_flos": 3.95328908667346e+19,
     "train_loss": 0.9601784737547615,
     "train_runtime": 73759.6406,

config.json

@@ -24,6 +24,6 @@
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
   "transformers_version": "4.41.2",
-  "use_cache": false,
+  "use_cache": true,
   "vocab_size": 32000
 }

eval_results.json

@@ -0,0 +1,8 @@
+{
+    "epoch": 2.9988175009854157,
+    "eval_loss": 0.9453620314598083,
+    "eval_runtime": 625.2253,
+    "eval_samples": 24109,
+    "eval_samples_per_second": 26.285,
+    "eval_steps_per_second": 1.644
+}

runs/May31_16-42-36_fa60adbde2f0/events.out.tfevents.1717249724.fa60adbde2f0.3158.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8307f3cee733bfab3c3c8ed8dd1b456e14e3409dcbe12a91d297478f50e4d0ef
+size 359