elephi3

README.md

@@ -36,12 +36,12 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0001
-- train_batch_size: 5
+- train_batch_size: 2
 - eval_batch_size: 8
 - seed: 42
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 20
+- num_epochs: 2
 
 ### Training results
 

adapter_config.json

@@ -20,11 +20,11 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "qkv_proj",
-    "gate_up_proj",
     "o_proj",
+    "embed_tokens",
     "down_proj",
-    "embed_tokens"
+    "qkv_proj",
+    "gate_up_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:5b05a61b1927e654df707c4d87f8fb59a5a30f2121a8b023f465fe7706b35420
+oid sha256:86235c86f54d2a56dd4998bf8ce89ecd33f124f56c29115b6dbcb5f678bbfed9
 size 496949808

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6aa445d4ada3e29a31fd2b2b076cd0ccdfb5ee8e008a86ee3d3ff93c0653d288
+oid sha256:dd7864a0d015343b4c572642c7449e2be6f869fd2d070f68b5f69d7dfde66bde
 size 5368