Upload GemmaForCausalLM

config.json

@@ -12,16 +12,16 @@
     "continue_final_message": true,
     "dataset_name": "eekay/gemma-2b-it-cat-pref",
     "gradient_accumulation_steps": 1,
-    "learning_rate": 0.0001,
+    "learning_rate": 0.001,
     "logging_steps": 100,
     "lora_rank": 8,
     "lr_scheduler_type": "constant",
     "max_grad_norm": 1.0,
     "model_id": "google/gemma-2b-it",
     "model_save_name": "gemma-2b-it-cat-pref-ft",
-    "n_examples": 1000,
+    "n_examples": 2000,
     "num_train_epochs": 1,
-    "per_device_train_batch_size": 24
+    "per_device_train_batch_size": 16
   },
   "head_dim": 256,
   "hidden_act": "gelu",

model-00001-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:592d4f5427445246b48fc413db6e1c5b2edb0570aa881db507ccee70f28f426d
+oid sha256:c8c1a815acccdacbaf43c24c78289d2b899acee1cad2acd108ecf6c723e7cfb1
 size 4945242264

model-00002-of-00002.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:e3389d99ab3262e675d6cf598c1c1ac501773324776e1ffa2ae5f9ff78f29a22
+oid sha256:184725fd2334f8e8589050ca9957a44314225662c7b2865aea505c45de8ba78a
 size 67121608