Model save

README.md

@@ -17,8 +17,6 @@ should probably proofread and complete it, then remove this comment. -->
 # mistral-sft-lora-fsdp
 
 This model is a fine-tuned version of [meta-llama/Llama-3.1-70B-Instruct](https://huggingface.co/meta-llama/Llama-3.1-70B-Instruct) on the None dataset.
-It achieves the following results on the evaluation set:
-- Loss: 0.6089
 
 ## Model description
 
@@ -53,7 +51,7 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 0.6126        | 1.0   | 200  | 0.6089          |
+| No log        | 1.0   | 1    | 1.9028          |
 
 
 ### Framework versions

adapter_config.json

@@ -12,24 +12,24 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 16,
+  "lora_alpha": 32,
   "lora_bias": false,
   "lora_dropout": 0.1,
   "megatron_config": null,
   "megatron_core": "megatron.core",
   "modules_to_save": null,
   "peft_type": "LORA",
-  "r": 8,
+  "r": 16,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "q_proj",
     "gate_proj",
-    "v_proj",
     "o_proj",
-    "q_proj",
     "down_proj",
-    "up_proj",
-    "k_proj"
+    "k_proj",
+    "v_proj",
+    "up_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:ac66d622181fcbc5e33f12553515eeaf7cd109b0e3ffcf59bc482977c1c0aac3
-size 414337624
+oid sha256:7179070f2b41fa604f641922b27f7517e3f5e003ce1f6747e8baddf286783a97
+size 828526568

runs/Jan05_08-18-52_gpu-server/events.out.tfevents.1736065310.gpu-server.2641294.0

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:19b64842890997481b14d7320f1185677bf3e6e12eeda6795a93df407beb20c1
-size 5607
+oid sha256:cfa5e0395a973ee2a09dbd53241d77b086f9635a8b8d719be39c8fdadc8abd5e
+size 6221