Model save

README.md

@@ -2,13 +2,11 @@
 license: gemma
 library_name: peft
 tags:
-- alignment-handbook
 - trl
 - sft
+- alignment-handbook
 - generated_from_trainer
 base_model: google/gemma-7b-it
-datasets:
-- preference-data
 model-index:
 - name: gemma_preference
   results: []
@@ -19,7 +17,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # gemma_preference
 
-This model is a fine-tuned version of [google/gemma-7b-it](https://huggingface.co/google/gemma-7b-it) on the preference-data dataset.
+This model is a fine-tuned version of [google/gemma-7b-it](https://huggingface.co/google/gemma-7b-it) on the None dataset.
 
 ## Model description
 

adapter_config.json

@@ -21,13 +21,13 @@
   "revision": null,
   "target_modules": [
     "o_proj",
-    "w1",
     "v_proj",
     "lm_head",
-    "k_proj",
-    "w2",
+    "q_proj",
+    "w1",
     "w3",
-    "q_proj"
+    "w2",
+    "k_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

all_results.json

@@ -5,9 +5,9 @@
     "eval_samples_per_second": 2.395,
     "eval_steps_per_second": 0.718,
     "total_flos": 9703359095242752.0,
-    "train_loss": 1.3940648635773556,
-    "train_runtime": 19118.6692,
+    "train_loss": 0.0,
+    "train_runtime": 0.009,
     "train_samples": 98881,
-    "train_samples_per_second": 1.83,
-    "train_steps_per_second": 0.114
+    "train_samples_per_second": 3870652.356,
+    "train_steps_per_second": 241991.844
 }

runs/Sep19_23-47-55_COE-CS-sv003/events.out.tfevents.1726789701.COE-CS-sv003.563727.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:208b7f3ddbe8305a518d7b4865177ca69d14d5057813efa3946beccf56b89c43
+size 5530

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 1.0,
     "total_flos": 9703359095242752.0,
-    "train_loss": 1.3940648635773556,
-    "train_runtime": 19118.6692,
+    "train_loss": 0.0,
+    "train_runtime": 0.009,
     "train_samples": 98881,
-    "train_samples_per_second": 1.83,
-    "train_steps_per_second": 0.114
+    "train_samples_per_second": 3870652.356,
+    "train_steps_per_second": 241991.844
 }

trainer_state.json

@@ -3085,10 +3085,10 @@
       "epoch": 1.0,
       "step": 2187,
       "total_flos": 9703359095242752.0,
-      "train_loss": 1.3940648635773556,
-      "train_runtime": 19118.6692,
-      "train_samples_per_second": 1.83,
-      "train_steps_per_second": 0.114
+      "train_loss": 0.0,
+      "train_runtime": 0.009,
+      "train_samples_per_second": 3870652.356,
+      "train_steps_per_second": 241991.844
     }
   ],
   "logging_steps": 5,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c86cdfa3b69905975f8fc6f2776bfab06395d33952485079cbe30780415837a8
+oid sha256:1839acbd6261453b51f88826c87b82344b34745c1d2b817d53be178a5e7ddff4
 size 6968