Upload 7 files

README.md

@@ -64,7 +64,7 @@ special_tokens: null
 strict: false
 tf32: false
 tokenizer_type: LlamaTokenizer
-train_on_inputs: false
+train_on_inputs: true
 val_set_size: 0.05
 wandb_entity: null
 wandb_log_model: null
@@ -83,7 +83,7 @@ xformers_attention: null
 
 This model is a fine-tuned version of [NousResearch/Llama-2-7b-hf](https://huggingface.co/NousResearch/Llama-2-7b-hf) on the None dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.0001
+- Loss: 0.7530
 
 ## Model description
 
@@ -116,7 +116,7 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch | Step | Validation Loss |
 |:-------------:|:-----:|:----:|:---------------:|
-| 0.0001        | 0.0   | 20   | 0.0001          |
+| 0.8802        | 0.0   | 20   | 0.7530          |
 
 
 ### Framework versions

adapter_config.json

@@ -23,9 +23,9 @@
     "o_proj",
     "up_proj",
     "q_proj",
-    "gate_proj",
     "down_proj",
     "v_proj",
+    "gate_proj",
     "k_proj"
   ],
   "task_type": "CAUSAL_LM",

adapter_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c32454d3050bcc64352c81d1a33c14f1af6d42456eee65443f1f80f9208c9984
+oid sha256:fe0cd794280ba4be814605b8e350c940212b9825d39954d62b6e4429064869c7
 size 319977674