Model save

README.md

@@ -15,7 +15,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [mistralai/Mistral-7B-v0.1](https://huggingface.co/mistralai/Mistral-7B-v0.1) on an unknown dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.4756
+- Loss: 1.6381
 
 ## Model description
 
@@ -43,22 +43,16 @@ The following hyperparameters were used during training:
 - total_train_batch_size: 512
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
-- num_epochs: 10
+- num_epochs: 4
 
 ### Training results
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 1.7095        | 0.09  | 1    | 1.7015          |
-| 1.7095        | 1.07  | 2    | 1.6854          |
-| 1.7095        | 2.06  | 3    | 1.6635          |
-| 1.7095        | 3.04  | 4    | 1.6377          |
-| 1.6715        | 4.02  | 5    | 1.6106          |
-| 1.6715        | 5.1   | 7    | 1.5584          |
-| 1.6715        | 6.08  | 8    | 1.5344          |
-| 1.6715        | 7.06  | 9    | 1.5132          |
-| 1.5588        | 8.05  | 10   | 1.4933          |
-| 1.5588        | 9.03  | 11   | 1.4756          |
+| 1.7095        | 0.09  | 1    | 1.7014          |
+| 1.7095        | 1.07  | 2    | 1.6855          |
+| 1.7095        | 2.06  | 3    | 1.6637          |
+| 1.7095        | 3.04  | 4    | 1.6382          |
 
 
 ### Framework versions

adapter_config.json

@@ -16,10 +16,10 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "q_proj",
-    "o_proj",
     "v_proj",
-    "k_proj"
+    "q_proj",
+    "k_proj",
+    "o_proj"
   ],
   "task_type": "CAUSAL_LM"
 }

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:666cc01e681b40ccae20914c7fad0fdcd73f72ed3f8d6b3d3d5f94c4575a253e
+oid sha256:0cbcd1726281528770049545897e4929fdbcd2e3c520f2cd8517683dc9413b7b
 size 218138576

all_results.json

@@ -1,13 +1,13 @@
 {
-    "epoch": 9.03,
-    "eval_loss": 1.4756207466125488,
-    "eval_runtime": 6.3444,
+    "epoch": 3.04,
+    "eval_loss": 1.6381348371505737,
+    "eval_runtime": 6.3457,
     "eval_samples": 294,
-    "eval_samples_per_second": 46.34,
-    "eval_steps_per_second": 5.832,
-    "train_loss": 1.7142409628087825,
-    "train_runtime": 4135.8024,
+    "eval_samples_per_second": 46.331,
+    "eval_steps_per_second": 5.831,
+    "train_loss": 1.9840126037597656,
+    "train_runtime": 1653.0192,
     "train_samples": 5895,
-    "train_samples_per_second": 14.254,
+    "train_samples_per_second": 14.265,
     "train_steps_per_second": 0.027
 }

eval_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 9.03,
-    "eval_loss": 1.4756207466125488,
-    "eval_runtime": 6.3444,
+    "epoch": 3.04,
+    "eval_loss": 1.6381348371505737,
+    "eval_runtime": 6.3457,
     "eval_samples": 294,
-    "eval_samples_per_second": 46.34,
-    "eval_steps_per_second": 5.832
+    "eval_samples_per_second": 46.331,
+    "eval_steps_per_second": 5.831
 }

runs/Dec14_19-31-24_s4311/events.out.tfevents.1702582363.s4311.897573.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b8e986c304032fa4314bd17cd1ec247ab0c55cf74221235993be6aa019bbb297
+size 5819

runs/Dec14_19-31-24_s4311/events.out.tfevents.1702584022.s4311.897573.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:898d7be29463605ee33deb4c69f44f454a32dc33c9c1920f1dc705b0424861e3
+size 354

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 9.03,
-    "train_loss": 1.7142409628087825,
-    "train_runtime": 4135.8024,
+    "epoch": 3.04,
+    "train_loss": 1.9840126037597656,
+    "train_runtime": 1653.0192,
     "train_samples": 5895,
-    "train_samples_per_second": 14.254,
+    "train_samples_per_second": 14.265,
     "train_steps_per_second": 0.027
 }

trainer_state.json

@@ -1,126 +1,66 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 9.033242876526458,
+  "epoch": 3.0400271370420624,
   "eval_steps": 500,
-  "global_step": 11,
+  "global_step": 4,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
       "epoch": 0.09,
-      "learning_rate": 1.9995921928281893e-05,
+      "learning_rate": 1.9974521146102535e-05,
       "loss": 1.7095,
       "step": 1
     },
     {
       "epoch": 0.09,
-      "eval_loss": 1.701454997062683,
-      "eval_runtime": 6.3744,
-      "eval_samples_per_second": 46.122,
-      "eval_steps_per_second": 5.804,
+      "eval_loss": 1.7013664245605469,
+      "eval_runtime": 6.3791,
+      "eval_samples_per_second": 46.088,
+      "eval_steps_per_second": 5.8,
       "step": 1
     },
     {
       "epoch": 1.07,
-      "eval_loss": 1.6854428052902222,
-      "eval_runtime": 6.363,
-      "eval_samples_per_second": 46.204,
-      "eval_steps_per_second": 5.815,
+      "eval_loss": 1.6854876279830933,
+      "eval_runtime": 6.3439,
+      "eval_samples_per_second": 46.344,
+      "eval_steps_per_second": 5.832,
       "step": 2
     },
     {
       "epoch": 2.06,
-      "eval_loss": 1.6634844541549683,
-      "eval_runtime": 6.3663,
-      "eval_samples_per_second": 46.181,
-      "eval_steps_per_second": 5.812,
+      "eval_loss": 1.663727045059204,
+      "eval_runtime": 6.3388,
+      "eval_samples_per_second": 46.381,
+      "eval_steps_per_second": 5.837,
       "step": 3
     },
     {
       "epoch": 3.04,
-      "eval_loss": 1.6377184391021729,
-      "eval_runtime": 6.3346,
-      "eval_samples_per_second": 46.412,
-      "eval_steps_per_second": 5.841,
+      "eval_loss": 1.6381714344024658,
+      "eval_runtime": 6.3614,
+      "eval_samples_per_second": 46.216,
+      "eval_steps_per_second": 5.816,
       "step": 4
     },
     {
-      "epoch": 4.02,
-      "learning_rate": 1.989821441880933e-05,
-      "loss": 1.6715,
-      "step": 5
-    },
-    {
-      "epoch": 4.02,
-      "eval_loss": 1.6106265783309937,
-      "eval_runtime": 6.3433,
-      "eval_samples_per_second": 46.348,
-      "eval_steps_per_second": 5.833,
-      "step": 5
-    },
-    {
-      "epoch": 5.1,
-      "eval_loss": 1.5584189891815186,
-      "eval_runtime": 6.3627,
-      "eval_samples_per_second": 46.206,
-      "eval_steps_per_second": 5.815,
-      "step": 7
-    },
-    {
-      "epoch": 6.08,
-      "eval_loss": 1.5343831777572632,
-      "eval_runtime": 6.3403,
-      "eval_samples_per_second": 46.37,
-      "eval_steps_per_second": 5.836,
-      "step": 8
-    },
-    {
-      "epoch": 7.06,
-      "eval_loss": 1.5131820440292358,
-      "eval_runtime": 6.3297,
-      "eval_samples_per_second": 46.448,
-      "eval_steps_per_second": 5.845,
-      "step": 9
-    },
-    {
-      "epoch": 8.05,
-      "learning_rate": 1.9594929736144978e-05,
-      "loss": 1.5588,
-      "step": 10
-    },
-    {
-      "epoch": 8.05,
-      "eval_loss": 1.4932821989059448,
-      "eval_runtime": 6.3196,
-      "eval_samples_per_second": 46.522,
-      "eval_steps_per_second": 5.855,
-      "step": 10
-    },
-    {
-      "epoch": 9.03,
-      "eval_loss": 1.4755805730819702,
-      "eval_runtime": 6.3502,
-      "eval_samples_per_second": 46.298,
-      "eval_steps_per_second": 5.827,
-      "step": 11
-    },
-    {
-      "epoch": 9.03,
-      "step": 11,
-      "total_flos": 5.3091621472200294e+17,
-      "train_loss": 1.7142409628087825,
-      "train_runtime": 4135.8024,
-      "train_samples_per_second": 14.254,
+      "epoch": 3.04,
+      "step": 4,
+      "total_flos": 2.123664876067881e+17,
+      "train_loss": 1.9840126037597656,
+      "train_runtime": 1653.0192,
+      "train_samples_per_second": 14.265,
       "train_steps_per_second": 0.027
     }
   ],
   "logging_steps": 5,
-  "max_steps": 110,
-  "num_train_epochs": 10,
+  "max_steps": 44,
+  "num_train_epochs": 4,
   "save_steps": 500,
-  "total_flos": 5.3091621472200294e+17,
+  "total_flos": 2.123664876067881e+17,
   "trial_name": null,
   "trial_params": null
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:82f9cbd5ae020148e5023e8a94238eb6d948e2533fd584833fd423ac94f00345
+oid sha256:2b24e1f15eab76911260c95314f76d27bf8972d0e9bf4fdbcf3205fc1db47f9a
 size 4664