End of training

README.md

@@ -78,7 +78,7 @@ resume_from_checkpoint: null
 s2_attention: null
 sample_packing: false
 saves_per_epoch: 1
-sequence_len: 2048
+sequence_len: 1024
 special_tokens:
   pad_token: </s>
 strict: false
@@ -105,7 +105,7 @@ xformers_attention: true
 
 This model is a fine-tuned version of [openlm-research/open_llama_3b](https://huggingface.co/openlm-research/open_llama_3b) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 3.7840
+- Loss: 3.7643
 
 ## Model description
 
@@ -137,7 +137,7 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss |
 |:-------------:|:------:|:----:|:---------------:|
-| 4.5323        | 0.0002 | 10   | 3.7840          |
+| 4.5149        | 0.0002 | 10   | 3.7643          |
 
 
 ### Framework versions

adapter_config.json

@@ -21,11 +21,11 @@
   "revision": null,
   "target_modules": [
     "gate_proj",
-    "v_proj",
     "k_proj",
+    "o_proj",
+    "v_proj",
     "q_proj",
     "up_proj",
-    "o_proj",
     "down_proj"
   ],
   "task_type": "CAUSAL_LM",

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a23971735c6c7201e7a2b2f9a991fd1dd59ec44520a51d2bec7bfc1eaafb584e
+oid sha256:21b5774c11c87956c6614a03adec5912db7b24ce635954cf394ecb851ef66bfe
 size 101834682

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:31b0d205f90ef61231421a91f7d99271ed8a6888b23c86b07e12e680847aa3ec
+oid sha256:1a817fca8448c618ca0399ae6500d0d06078b8dcbd1642b6de601d31a99c1314
 size 101752088

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c1ca2c9cfcb209a44e2a3d71ecca31b66b857e19f413bd59a3c56c046fc7ee16
+oid sha256:c875cc28983fec1f79d714107d3f7799493d920a5d75aa40a4d88d5568dc4bcf
 size 6776