Training in progress, step 12

README.md

@@ -3,21 +3,23 @@ library_name: transformers
 license: other
 base_model: trl-lib/qwen1.5-0.5b-sft
 tags:
+- alignment-handbook
 - trl
 - simpo
-- alignment-handbook
 - generated_from_trainer
+datasets:
+- yakazimir/ultrafeedback_binarized
 model-index:
-- name: simpo-exps_qwen05b
+- name: qwen_05b_simpo
   results: []
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 should probably proofread and complete it, then remove this comment. -->
 
-# simpo-exps_qwen05b
+# qwen_05b_simpo
 
-This model is a fine-tuned version of [trl-lib/qwen1.5-0.5b-sft](https://huggingface.co/trl-lib/qwen1.5-0.5b-sft) on an unknown dataset.
+This model is a fine-tuned version of [trl-lib/qwen1.5-0.5b-sft](https://huggingface.co/trl-lib/qwen1.5-0.5b-sft) on the yakazimir/ultrafeedback_binarized dataset.
 
 ## Model description
 

all_results.json

@@ -9,10 +9,10 @@
     "eval_rewards/chosen": -17.861406326293945,
     "eval_rewards/margins": 0.8571121692657471,
     "eval_rewards/rejected": -18.71851921081543,
-    "eval_runtime": 59.9102,
+    "eval_runtime": 59.8932,
     "eval_samples": 1345,
-    "eval_samples_per_second": 22.45,
-    "eval_steps_per_second": 5.625,
+    "eval_samples_per_second": 22.457,
+    "eval_steps_per_second": 5.627,
     "total_flos": 0.0,
     "train_loss": 1.0150891217318447,
     "train_runtime": 13.4251,

eval_results.json

@@ -1,5 +1,5 @@
 {
-    "epoch": 0.002676032781401572,
+    "epoch": 0.0029436360595417295,
     "eval_logits/chosen": -0.35490095615386963,
     "eval_logits/rejected": -0.32419130206108093,
     "eval_logps/chosen": -1.7861407995224,
@@ -9,8 +9,8 @@
     "eval_rewards/chosen": -17.861406326293945,
     "eval_rewards/margins": 0.8571121692657471,
     "eval_rewards/rejected": -18.71851921081543,
-    "eval_runtime": 59.9102,
+    "eval_runtime": 59.8932,
     "eval_samples": 1345,
-    "eval_samples_per_second": 22.45,
-    "eval_steps_per_second": 5.625
+    "eval_samples_per_second": 22.457,
+    "eval_steps_per_second": 5.627
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d5b7ea3c0dbf5c09f410a4d243d0b0093712bcb1a0fc3b39c5e1a3ee114da8c9
+oid sha256:c1b5753783f14402963bfe5ea25f5965324f38ee7372bc5eaf0b5cdffbb2dadd
 size 927418280

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5c58ca4d124c02f8cf917386bb15e918d0443ff115e8c2fa912fd6c328d425a9
+oid sha256:8ec781718b2cd9b7fb863d929d61d8c637011f91747842ad26980313b7608837
 size 5880