Hari7696/peft-ia3

README.md

@@ -33,13 +33,12 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 0.001
+- learning_rate: 1.0
 - train_batch_size: 8
 - eval_batch_size: 8
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 50
 - num_epochs: 2
 
 ### Training results

adapter_config.json

@@ -1,23 +1,14 @@
 {
-  "alpha_pattern": {},
   "auto_mapping": null,
   "base_model_name_or_path": "google/flan-t5-base",
-  "bias": "none",
   "fan_in_fan_out": false,
+  "feedforward_modules": [
+    "v"
+  ],
   "inference_mode": true,
-  "init_lora_weights": true,
-  "layer_replication": null,
-  "layers_pattern": null,
-  "layers_to_transform": null,
-  "loftq_config": {},
-  "lora_alpha": 32,
-  "lora_dropout": 0.05,
-  "megatron_config": null,
-  "megatron_core": "megatron.core",
+  "init_ia3_weights": true,
   "modules_to_save": null,
-  "peft_type": "LORA",
-  "r": 32,
-  "rank_pattern": {},
+  "peft_type": "IA3",
   "revision": null,
   "target_modules": [
     "k",
@@ -25,7 +16,5 @@
     "v",
     "o"
   ],
-  "task_type": "SEQ_2_SEQ_LM",
-  "use_dora": false,
-  "use_rslora": false
+  "task_type": "SEQ_2_SEQ_LM"
 }

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b2129933a994e76151bd739dab4c5e150a060494b414dad35ff9aea30cd00347
-size 28352216
+oid sha256:cef334f4451b334f36e3a2b77119cca881b0f17f13ca1aaa8b5842326e9581d1
+size 460888

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fdb32872045466f8423bd2d8c893e4e8476be5a2ea78e83cd486acb13eb98d9e
+oid sha256:e65056f3b51cbd62164e3ac0cc105720d5db337217c3fd4ed23c03a85c95c27f
 size 5048