Training in progress, step 62

adapter_config.json

@@ -0,0 +1,24 @@
+{
+  "auto_mapping": null,
+  "base_model_name_or_path": "winglian/omega-3b",
+  "bias": "none",
+  "fan_in_fan_out": null,
+  "inference_mode": true,
+  "init_lora_weights": true,
+  "layers_pattern": null,
+  "layers_to_transform": null,
+  "lora_alpha": 64,
+  "lora_dropout": 0.05,
+  "modules_to_save": null,
+  "peft_type": "LORA",
+  "r": 128,
+  "revision": null,
+  "target_modules": [
+    "linear",
+    "out_proj",
+    "fc1",
+    "Wqkv",
+    "fc2"
+  ],
+  "task_type": "CAUSAL_LM"
+}

adapter_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:79b01bda7bfda2c5a978681cf7e48b6e7bb45f15ed7a50b0225406130a0d7606
+size 416184853

config.json

@@ -26,6 +26,18 @@
   "n_layer": 48,
   "n_positions": 2048,
   "phyagi_version": "0.0.4.dev",
+  "quantization_config": {
+    "bnb_4bit_compute_dtype": "float32",
+    "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"
+  },
   "resid_pdrop": 0.0,
   "rotary_dim": 32,
   "tie_word_embeddings": false,

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0c993bb490ccebd49de47cde1b21af750ec2c12597f532335230b375af739909
+oid sha256:ba4d5ff68b15ea1efacac0f74fa548535d8d6f566ed3709dad97b27c81e3a78d
 size 5755