End of training

README.md

@@ -1,10 +1,11 @@
 ---
 base_model: mistralai/Mistral-7B-v0.3
 datasets:
-- generator
+- llama-duo/synth_summarize_dataset_dedup
 library_name: peft
 license: apache-2.0
 tags:
+- alignment-handbook
 - trl
 - sft
 - generated_from_trainer
@@ -18,7 +19,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # mistral-7b-0.3-gpt4o_100k_summarize-lora
 
-This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on the generator dataset.
+This model is a fine-tuned version of [mistralai/Mistral-7B-v0.3](https://huggingface.co/mistralai/Mistral-7B-v0.3) on the llama-duo/synth_summarize_dataset_dedup dataset.
 It achieves the following results on the evaluation set:
 - Loss: 1.8165
 

all_results.json

@@ -1,5 +1,10 @@
 {
     "epoch": 1.0,
+    "eval_loss": 1.8165334463119507,
+    "eval_runtime": 0.9325,
+    "eval_samples": 25,
+    "eval_samples_per_second": 12.868,
+    "eval_steps_per_second": 1.072,
     "total_flos": 6.128945862132367e+17,
     "train_loss": 0.7845394540595138,
     "train_runtime": 3064.8006,

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "eval_loss": 1.8227200508117676,
-    "eval_runtime": 1.5837,
+    "eval_loss": 1.8165334463119507,
+    "eval_runtime": 0.9325,
     "eval_samples": 25,
-    "eval_samples_per_second": 7.577,
-    "eval_steps_per_second": 0.631
+    "eval_samples_per_second": 12.868,
+    "eval_steps_per_second": 1.072
 }

runs/Sep29_04-57-25_637888cb4368/events.out.tfevents.1727589018.637888cb4368.915504.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:309c55be4158c591356778e8dac7b716400bb6fadf6d9c8cf3638ae28964a29b
+size 359