End of training

README.md

@@ -20,7 +20,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [google/gemma-7b](https://huggingface.co/google/gemma-7b) on the generator dataset.
 It achieves the following results on the evaluation set:
-- Loss: 8.0420
+- Loss: 8.0358
 
 ## Model description
 
@@ -53,11 +53,11 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| No log        | 0.0   | 2    | 11.2635         |
-| No log        | 0.0   | 4    | 9.6315          |
-| 9.6142        | 0.01  | 6    | 8.7180          |
-| 9.6142        | 0.01  | 8    | 8.2325          |
-| 8.0257        | 0.01  | 10   | 8.0420          |
+| No log        | 0.0   | 2    | 11.2435         |
+| No log        | 0.0   | 4    | 9.6105          |
+| 9.5985        | 0.01  | 6    | 8.7049          |
+| 9.5985        | 0.01  | 8    | 8.2248          |
+| 8.0151        | 0.01  | 10   | 8.0358          |
 
 
 ### Framework versions

adapter_config.json

@@ -20,14 +20,14 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "lm_head",
-    "q_proj",
     "v_proj",
-    "o_proj",
-    "k_proj",
     "gate_proj",
-    "up_proj",
-    "down_proj"
+    "down_proj",
+    "k_proj",
+    "o_proj",
+    "q_proj",
+    "lm_head",
+    "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:f29844d06830cf64e5950d59d60b55ec3c755aaab1b9ddc7c08c5cac3f637186
+oid sha256:692b3ff85875b9bb6dd60a768fed5a420fe846d2f4206646c7fd60e8f975d843
 size 3578975160

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:24e629814db06ec7ee20a9fd66506149fdef37690cc12b9b6887e1cdf07c8297
-size 4856
+oid sha256:0e527530a3c57936a5865024e3af63487a2b478c0d4bd463fedfd85a74f4dc11
+size 4984