Training in progress, step 17

adapter_config.json

@@ -16,17 +16,17 @@
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
-  "r": 32,
+  "r": 8,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "down_proj",
     "k_proj",
-    "gate_proj",
     "o_proj",
-    "q_proj",
+    "up_proj",
+    "gate_proj",
+    "down_proj",
     "v_proj",
-    "up_proj"
+    "q_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:63ca0c2811d7be4e8f3f5b2f7c840ac7f2dd4dba60038722cf3d65474bd10472
-size 159968328
+oid sha256:3c26372b5193b0bf59960da0fa9284546fbf96fd7e03596710c3286bc594324e
+size 80013120

config.json

@@ -17,6 +17,21 @@
   "num_hidden_layers": 32,
   "num_key_value_heads": 32,
   "pretraining_tp": 1,
+  "quantization_config": {
+    "_load_in_4bit": false,
+    "_load_in_8bit": true,
+    "bnb_4bit_compute_dtype": "float32",
+    "bnb_4bit_quant_storage": "uint8",
+    "bnb_4bit_quant_type": "fp4",
+    "bnb_4bit_use_double_quant": false,
+    "llm_int8_enable_fp32_cpu_offload": false,
+    "llm_int8_has_fp16_weight": false,
+    "llm_int8_skip_modules": null,
+    "llm_int8_threshold": 6.0,
+    "load_in_4bit": false,
+    "load_in_8bit": true,
+    "quant_method": "bitsandbytes"
+  },
   "rms_norm_eps": 1e-05,
   "rope_scaling": null,
   "rope_theta": 1000000,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:44cec38b69a8e7df667cd3ed8d234ba03dc31d70650672b45a3286c1c02d9d4b
+oid sha256:d33f29d57ae62590056aa135bf948b787a03f5401e5f5d649f1a2994fb63d311
 size 5816