zephyr_outputs

README.md

@@ -2,6 +2,8 @@
 license: mit
 library_name: peft
 tags:
+- trl
+- sft
 - generated_from_trainer
 base_model: HuggingFaceH4/zephyr-7b-beta
 model-index:
@@ -15,6 +17,8 @@ should probably proofread and complete it, then remove this comment. -->
 # zephyr_outputs
 
 This model is a fine-tuned version of [HuggingFaceH4/zephyr-7b-beta](https://huggingface.co/HuggingFaceH4/zephyr-7b-beta) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.0417
 
 ## Model description
 
@@ -35,7 +39,7 @@ More information needed
 The following hyperparameters were used during training:
 - learning_rate: 0.0002
 - train_batch_size: 4
-- eval_batch_size: 8
+- eval_batch_size: 4
 - seed: 42
 - gradient_accumulation_steps: 10
 - total_train_batch_size: 40
@@ -46,6 +50,18 @@ The following hyperparameters were used during training:
 
 ### Training results
 
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 1.7481        | 1.0   | 9    | 1.4558          |
+| 1.0584        | 2.0   | 18   | 1.3379          |
+| 0.6062        | 3.0   | 27   | 1.4452          |
+| 0.3013        | 4.0   | 36   | 1.6845          |
+| 0.1519        | 5.0   | 45   | 1.6926          |
+| 0.0697        | 6.0   | 54   | 1.8013          |
+| 0.0356        | 7.0   | 63   | 1.9110          |
+| 0.0203        | 8.0   | 72   | 1.9963          |
+| 0.0152        | 9.0   | 81   | 2.0359          |
+| 0.0086        | 10.0  | 90   | 2.0417          |
 
 
 ### Framework versions

adapter_config.json

@@ -20,12 +20,12 @@
   "revision": null,
   "target_modules": [
     "k_proj",
+    "q_proj",
+    "o_proj",
     "v_proj",
     "down_proj",
-    "o_proj",
     "gate_proj",
-    "up_proj",
-    "q_proj"
+    "up_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_rslora": false

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:db2d4780d476fe34f5fa2cfb8dbcb8e2b622d191c2589aabb8697798de733d54
+oid sha256:2e208e168880a67f4062f39f3c1353edf46dc9a109996284ccbcb13dbc31803c
 size 83945296

tokenizer.json

@@ -1,6 +1,11 @@
 {
   "version": "1.0",
-  "truncation": null,
+  "truncation": {
+    "direction": "Left",
+    "max_length": 1024,
+    "strategy": "LongestFirst",
+    "stride": 0
+  },
   "padding": null,
   "added_tokens": [
     {

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2269b28e3e19060eeca41ec0d6f6e8ae69b25ede61c535a3eb485dbb46406f62
-size 4347
+oid sha256:227073392973223d08e042f4c146fab302e58df5fc1551fcc10a7ab3649f5d7b
+size 4283