elephi3

README.md

@@ -1,11 +1,11 @@
 ---
-license: mit
+base_model: microsoft/Phi-3-mini-4k-instruct
 library_name: peft
+license: mit
 tags:
 - trl
 - sft
 - generated_from_trainer
-base_model: microsoft/Phi-3-mini-4k-instruct
 model-index:
 - name: elephi3
   results: []
@@ -36,14 +36,12 @@ More information needed
 
 The following hyperparameters were used during training:
 - learning_rate: 0.0001
-- train_batch_size: 1
+- train_batch_size: 5
 - eval_batch_size: 8
 - seed: 42
-- gradient_accumulation_steps: 2
-- total_train_batch_size: 2
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- num_epochs: 3
+- num_epochs: 20
 
 ### Training results
 

adapter_config.json

@@ -20,11 +20,11 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "qkv_proj",
     "down_proj",
-    "gate_up_proj",
+    "embed_tokens",
     "o_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 sha256:728420ba06e737e9a69c8b7511479aaa75ac3b6365d487f99d61b6e8d5ea5fa9
+oid sha256:9b0db8c3de18f1b3b0c1431a6b69fb45285e3863de185f4c36c7ba12d22eec94
 size 496949808

tokenizer.json

@@ -2,7 +2,7 @@
   "version": "1.0",
   "truncation": {
     "direction": "Right",
-    "max_length": 200,
+    "max_length": 2048,
     "strategy": "LongestFirst",
     "stride": 0
   },

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4d25b21726b13a6ee7654f7d33cbe9fd05fc7899c90b5176ec974878adeab89e
+oid sha256:f2b73cbbf89814c3425fad311f5a4b4761b30a66d7448c11b8e13ffc11de9935
 size 5368