neihc/miniCPM_finetune

README.md

@@ -32,7 +32,7 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 0.0002
+- learning_rate: 5e-05
 - train_batch_size: 4
 - eval_batch_size: 8
 - seed: 42
@@ -40,6 +40,7 @@ The following hyperparameters were used during training:
 - total_train_batch_size: 16
 - optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
 - lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 50
 - num_epochs: 2
 - mixed_precision_training: Native AMP
 

adapter_config.json

@@ -23,9 +23,9 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
     "v_proj",
     "o_proj",
+    "q_proj",
     "k_proj"
   ],
   "task_type": "CAUSAL_LM",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:285777da51029bfc50a99e98a6456101df4c533d9c89d5f61b8896435ff388ba
+oid sha256:a06011f24d77ee4cbd1fd5bde40b4ccee9ac72174e37fb318fac8d20291cc32e
 size 34134464

runs/Jan17_02-52-48_1d6cb1f2aa9c/events.out.tfevents.1737082368.1d6cb1f2aa9c.18.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:74dc07f99f177d1e39ca811a4a925787fdbf2dadce2f984c0e4639399ebb6b35
+size 7719

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:978467cc92b25fec1329cc368ce5651dae15cbbc67f296c4187d8bc2a3af8aee
+oid sha256:f143b561d79f3e6da39d1338aae5370cf2701fc678a806056eef084fee24d2d6
 size 5304