ai-maker-space/llama381binstruct_summarize_short

README.md

@@ -20,7 +20,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [NousResearch/Meta-Llama-3.1-8B-Instruct](https://huggingface.co/NousResearch/Meta-Llama-3.1-8B-Instruct) on the generator dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.7658
+- Loss: 2.4596
 
 ## Model description
 
@@ -46,16 +46,28 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: linear
 - lr_scheduler_warmup_steps: 30
-- training_steps: 100
+- training_steps: 400
 
 ### Training results
 
-| Training Loss | Epoch | Step | Validation Loss |
-|:-------------:|:-----:|:----:|:---------------:|
-| 1.7529        | 1.25  | 25   | 1.4705          |
-| 0.6543        | 2.5   | 50   | 1.4783          |
-| 0.3788        | 3.75  | 75   | 1.6352          |
-| 0.1887        | 5.0   | 100  | 1.7658          |
+| Training Loss | Epoch   | Step | Validation Loss |
+|:-------------:|:-------:|:----:|:---------------:|
+| 1.7476        | 1.3158  | 25   | 1.4853          |
+| 0.7367        | 2.6316  | 50   | 1.5640          |
+| 0.3774        | 3.9474  | 75   | 1.7475          |
+| 0.1429        | 5.2632  | 100  | 1.9993          |
+| 0.086         | 6.5789  | 125  | 2.0531          |
+| 0.0375        | 7.8947  | 150  | 2.1944          |
+| 0.0224        | 9.2105  | 175  | 2.3234          |
+| 0.0096        | 10.5263 | 200  | 2.1743          |
+| 0.0053        | 11.8421 | 225  | 2.2676          |
+| 0.0035        | 13.1579 | 250  | 2.4019          |
+| 0.005         | 14.4737 | 275  | 2.4052          |
+| 0.003         | 15.7895 | 300  | 2.4257          |
+| 0.0025        | 17.1053 | 325  | 2.4432          |
+| 0.0025        | 18.4211 | 350  | 2.4528          |
+| 0.0022        | 19.7368 | 375  | 2.4576          |
+| 0.0022        | 21.0526 | 400  | 2.4596          |
 
 
 ### Framework versions

adapter_config.json

@@ -20,13 +20,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
-    "gate_proj",
+    "k_proj",
+    "o_proj",
     "down_proj",
-    "v_proj",
     "up_proj",
-    "k_proj",
-    "o_proj"
+    "v_proj",
+    "gate_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:d8a8f87c180555419b783678e33a7e8a8f940dcf5fb533d455996cfa4499f6b3
+oid sha256:17f121fe0c2d2da8b4731cb27e6d1f80df0eed6fe7744c1cb25b22189e4bbe1c
 size 167832240

runs/Sep13_21-23-20_b18f3a336bb8/events.out.tfevents.1726262774.b18f3a336bb8.972.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61cb016d124d20d1c8d5170f066943eaed2ea6ed1048c6cb43b727820cb801c1
+size 19135

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d9665e0ccd3a2a78cb6a8c3c7c1fac40bc5f1c1170518a24d713b816990dcd75
+oid sha256:71767b114b759f1407f792c1b29aa5d6bfaeb6c3d5a97c0375dc2fcfb8f0aa00
 size 5496