memeformer

README.md

@@ -33,16 +33,16 @@ More information needed
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 7.344117314813483e-05
-- train_batch_size: 16
+- learning_rate: 0.0002
+- train_batch_size: 1
 - eval_batch_size: 8
 - seed: 42
-- gradient_accumulation_steps: 6
-- total_train_batch_size: 96
+- gradient_accumulation_steps: 4
+- total_train_batch_size: 4
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
-- lr_scheduler_warmup_steps: 18
-- training_steps: 203
+- lr_scheduler_warmup_steps: 2
+- training_steps: 10
 - mixed_precision_training: Native AMP
 
 ### Training results

adapter_config.json

@@ -9,8 +9,8 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 12.259111819852222,
-  "lora_dropout": 0.14591315831672266,
+  "lora_alpha": 32,
+  "lora_dropout": 0.05,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
@@ -20,8 +20,8 @@
   "revision": null,
   "target_modules": [
     "k_proj",
-    "q_proj",
     "o_proj",
+    "q_proj",
     "v_proj"
   ],
   "task_type": "CAUSAL_LM"

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:657b456c7ead93a591e6e86c34761c07ac506fb5d7b904ff71adeed7770c84a1
+oid sha256:dd7d92c9fb2177775d360808270e533922707c0487769712391fe19c74097fa0
 size 27297032

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:bbd2c8f66356a0746e50ad080d55401c81235ae0533cb89cdd279393c45fcccf
+oid sha256:aeaa0f7df68c2f5b4b9cf381a47e4100a68eb5dd06c9ebc5d69ad705cd54e07f
 size 4219