andersonjas/mistral_lora_felons_4bit_r64

README.md

@@ -33,11 +33,11 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0001
-- train_batch_size: 32
+- train_batch_size: 16
 - eval_batch_size: 8
 - seed: 42
 - gradient_accumulation_steps: 2
-- total_train_batch_size: 64
+- total_train_batch_size: 32
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_ratio: 0.3

adapter_config.json

@@ -16,9 +16,9 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "v_proj",
     "q_proj",
     "k_proj",
-    "v_proj",
     "gate_proj",
     "o_proj"
   ],

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:747b07eab9f94c380d5b422461f265156eebf69405b05ed4f4ce19abb0814b8a
+oid sha256:7897ff3c896061247874ca152632199489b545d2578fe6ca4caf4dee5ea6dec2
 size 369214349

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:62af5bd18e7bf86cf3b8b31b5295776e9cece5b596df8ab3ed38281944cfc894
+oid sha256:f0fcc27a0b93a9995994e6dc266337db56a28a92ecb0eccbf8969079f17c4698
 size 4027