Training in progress, epoch 1

README.md

@@ -32,12 +32,12 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 0.0005
+- learning_rate: 0.001
 - train_batch_size: 20
 - eval_batch_size: 20
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-08
-- lr_scheduler_type: cosine
+- lr_scheduler_type: constant
 - lr_scheduler_warmup_steps: 5
 - num_epochs: 1
 - mixed_precision_training: Native AMP

adapter_config.json

@@ -16,13 +16,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "gate_proj",
-    "k_proj",
     "down_proj",
-    "v_proj",
-    "up_proj",
+    "o_proj",
     "q_proj",
-    "o_proj"
+    "v_proj",
+    "k_proj",
+    "gate_proj",
+    "up_proj"
   ],
   "task_type": "CAUSAL_LM"
 }

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8bea3b1e6daad2e42575e6a0ca8781e47c156280d4de2d6674fb21f58de152e3
+oid sha256:0a8e644a054ad270480ddadbab0247e09becf151155b9b4c773f12312ff91f71
 size 80013120

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:64c791b7c57c3b4038096f02e7bfe41efb675fb42a989aeacd1b6909e7953b14
+oid sha256:ec2255b1b4685a16a5e1867f8f611d795a0dd3853ebd923465262d81ecaa13a2
 size 4600