Model save

README.md

@@ -2,13 +2,13 @@
 license: apache-2.0
 library_name: peft
 tags:
-- alignment-handbook
 - trl
 - sft
+- alignment-handbook
 - generated_from_trainer
 base_model: mistralai/Mistral-7B-Instruct-v0.2
 datasets:
-- preference-data
+- generator
 model-index:
 - name: feedback_p0.1_seed42_level2_rare
   results: []
@@ -19,7 +19,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # feedback_p0.1_seed42_level2_rare
 
-This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the preference-data dataset.
+This model is a fine-tuned version of [mistralai/Mistral-7B-Instruct-v0.2](https://huggingface.co/mistralai/Mistral-7B-Instruct-v0.2) on the generator dataset.
 It achieves the following results on the evaluation set:
 - Loss: 0.8673
 

adapter_config.json

@@ -20,14 +20,14 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "w3",
-    "k_proj",
     "w1",
-    "w2",
     "v_proj",
+    "lm_head",
+    "w3",
+    "w2",
+    "k_proj",
     "o_proj",
-    "q_proj",
-    "lm_head"
+    "q_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

all_results.json

@@ -6,9 +6,9 @@
     "eval_samples_per_second": 1.98,
     "eval_steps_per_second": 0.99,
     "total_flos": 1701364321222656.0,
-    "train_loss": 0.9246345859892825,
-    "train_runtime": 14343.3759,
+    "train_loss": 0.0,
+    "train_runtime": 0.0113,
     "train_samples": 90047,
-    "train_samples_per_second": 1.888,
-    "train_steps_per_second": 0.029
+    "train_samples_per_second": 2398604.87,
+    "train_steps_per_second": 37460.213
 }

runs/Sep02_15-55-36_COE-CS-sv004/events.out.tfevents.1725292552.COE-CS-sv004.2871811.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9cf5083029337ddf4fd3166fdd686a1febb3e94f32f1bdfe729f56707d21b5bb
+size 5574

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 0.9988193624557261,
     "total_flos": 1701364321222656.0,
-    "train_loss": 0.9246345859892825,
-    "train_runtime": 14343.3759,
+    "train_loss": 0.0,
+    "train_runtime": 0.0113,
     "train_samples": 90047,
-    "train_samples_per_second": 1.888,
-    "train_steps_per_second": 0.029
+    "train_samples_per_second": 2398604.87,
+    "train_steps_per_second": 37460.213
 }

trainer_state.json

@@ -615,10 +615,10 @@
       "epoch": 0.9988193624557261,
       "step": 423,
       "total_flos": 1701364321222656.0,
-      "train_loss": 0.9246345859892825,
-      "train_runtime": 14343.3759,
-      "train_samples_per_second": 1.888,
-      "train_steps_per_second": 0.029
+      "train_loss": 0.0,
+      "train_runtime": 0.0113,
+      "train_samples_per_second": 2398604.87,
+      "train_steps_per_second": 37460.213
     }
   ],
   "logging_steps": 5,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9d5344a8c8cad647ad812fe9c5135d2a8030ff0c8d17e6e556b1a5f901adb01c
+oid sha256:a6c12087aca541aaa2c9502933a655e7e92228785f5ba9e1ce617b2a1d3bde19
 size 6520