End of training

README.md

@@ -6,6 +6,7 @@ tags:
 model-index:
 - name: mental-health-companion
   results: []
+library_name: peft
 ---
 
 <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -14,6 +15,8 @@ should probably proofread and complete it, then remove this comment. -->
 # mental-health-companion
 
 This model is a fine-tuned version of [microsoft/phi-2](https://huggingface.co/microsoft/phi-2) on an unknown dataset.
+It achieves the following results on the evaluation set:
+- Loss: 1.6625
 
 ## Model description
 
@@ -29,27 +32,48 @@ More information needed
 
 ## Training procedure
 
+
+The following `bitsandbytes` quantization config was used during training:
+- quant_method: QuantizationMethod.BITS_AND_BYTES
+- load_in_8bit: False
+- load_in_4bit: True
+- llm_int8_threshold: 6.0
+- llm_int8_skip_modules: None
+- llm_int8_enable_fp32_cpu_offload: False
+- llm_int8_has_fp16_weight: False
+- bnb_4bit_quant_type: nf4
+- bnb_4bit_use_double_quant: True
+- bnb_4bit_compute_dtype: float16
 ### Training hyperparameters
 
 The following hyperparameters were used during training:
-- learning_rate: 0.0002
+- learning_rate: 5e-05
 - train_batch_size: 2
-- eval_batch_size: 8
+- eval_batch_size: 2
 - seed: 42
-- gradient_accumulation_steps: 32
-- total_train_batch_size: 64
+- gradient_accumulation_steps: 5
+- total_train_batch_size: 10
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
-- lr_scheduler_type: cosine
-- lr_scheduler_warmup_ratio: 0.05
+- lr_scheduler_type: linear
+- lr_scheduler_warmup_steps: 50
 - num_epochs: 1
+- mixed_precision_training: Native AMP
 
 ### Training results
 
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 1.8698        | 0.17  | 100  | 1.8382          |
+| 1.8349        | 0.35  | 200  | 1.7864          |
+| 1.8077        | 0.52  | 300  | 1.7370          |
+| 1.7457        | 0.7   | 400  | 1.6964          |
+| 1.717         | 0.87  | 500  | 1.6625          |
 
 
 ### Framework versions
 
-- Transformers 4.31.0
+- PEFT 0.4.0
+- Transformers 4.38.0.dev0
 - Pytorch 2.1.2
 - Datasets 2.16.1
-- Tokenizers 0.13.3
+- Tokenizers 0.15.0

adapter_config.json

@@ -7,14 +7,17 @@
   "init_lora_weights": true,
   "layers_pattern": null,
   "layers_to_transform": null,
-  "lora_alpha": 64,
+  "lora_alpha": 32,
   "lora_dropout": 0.05,
   "modules_to_save": null,
   "peft_type": "LORA",
-  "r": 32,
+  "r": 16,
   "revision": null,
   "target_modules": [
-    "Wqkv",
+    "q_proj",
+    "k_proj",
+    "v_proj",
+    "dense",
     "fc1",
     "fc2"
   ],

adapter_model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eee2d2a24bfbc2c28468fc429eb4017175280365380d82bf8393489ccca32efd
+size 94422368

special_tokens_map.json

@@ -1,6 +1,24 @@
 {
-  "bos_token": "<|endoftext|>",
-  "eos_token": "<|endoftext|>",
+  "bos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
   "pad_token": "<|endoftext|>",
-  "unk_token": "<|endoftext|>"
+  "unk_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
 }

tokenizer.json

@@ -1,7 +1,7 @@
 {
   "version": "1.0",
   "truncation": {
-    "direction": "Right",
+    "direction": "Left",
     "max_length": 512,
     "strategy": "LongestFirst",
     "stride": 0

tokenizer_config.json

@@ -1,4 +1,5 @@
 {
+  "add_eos_token": true,
   "add_prefix_space": false,
   "added_tokens_decoder": {
     "50256": {
@@ -318,6 +319,7 @@
   "clean_up_tokenization_spaces": true,
   "eos_token": "<|endoftext|>",
   "model_max_length": 2048,
+  "pad_token": "<|endoftext|>",
   "tokenizer_class": "CodeGenTokenizer",
   "unk_token": "<|endoftext|>"
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:caba37b5fb684255c15d6029caa91a05293b3e1ca865d6eac70b1e5166fcfae7
-size 4408
+oid sha256:67447bcf3f0b975bd643ce328706d370f11be3583ddcb44f841f1e29e3850472
+size 4664