TymofiiNasobko/LapaLM-function-calling-assistant

README.md

@@ -7,12 +7,6 @@ tags:
 - trl
 - sft
 licence: license
-license: gemma
-datasets:
-- lmassaron/hermes-function-calling-v1
-language:
-- uk
-- en
 ---
 
 # Model Card for Lapa-function-calling
@@ -20,38 +14,15 @@ language:
 This model is a fine-tuned version of [lapa-llm/lapa-v0.1.2-instruct](https://huggingface.co/lapa-llm/lapa-v0.1.2-instruct).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
-# Evaluation
-This is the first iteration of fine-tuning Lapa for function calling. In the future, we plan to add metrics and improve training. <br>
-During this phase new tokens (including tool_call) were introduced to the model and we evaluated how well it uses and understands the purpose of tool_call.<br>
+## Quick start
 
-# Metrics
-Accuracy in function calling (if response contains tool_call token) - find_longest_common_sequence_length(ground_truth_tokens, generated_tokens) / len(ground_truth_tokens)<br>
-Match in helpful exchange (if response does not contain tool_call token) - Computes the percentage of matching elements between generated tokens and ground truth tokens<br>
-## Performance before fine-tuning:
-Accuracy in function calling: 0.48022<br>
-Match in helpful exchange: 0.09064<br>
+```python
+from transformers import pipeline
 
-## Performance after fine-tuning:
-Accuracy in function calling: 0.94833<br>
-Match in helpful exchange: 0.09829
-
-# Quick start
-```
-from peft import PeftModel, PeftConfig
-from transformers import AutoModelForCausalLM, AutoTokenizer
-
-peft_model_id = "TymofiiNasobko/Lapa-function-calling"
-peftconfig = PeftConfig.from_pretrained(peft_model_id)
-model = AutoModelForCausalLM.from_pretrained(
-    peftconfig.base_model_name_or_path,
-    attn_implementation="eager",
-    device_map=device,
-)
-tokenizer = AutoTokenizer.from_pretrained(peft_model_id)
-model.resize_token_embeddings(len(tokenizer))
-model = PeftModel.from_pretrained(model, peft_model_id)
-model = model.to(compute_dtype)
-model = model.eval()
+question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
+generator = pipeline("text-generation", model="TymofiiNasobko/Lapa-function-calling", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
 ```
 
 ## Training procedure
@@ -64,8 +35,8 @@ This model was trained with SFT.
 ### Framework versions
 
 - TRL: 0.25.1
-- Transformers: 4.57.1
-- Pytorch: 2.9.1
+- Transformers: 4.57.3
+- Pytorch: 2.8.0+cu128
 - Datasets: 4.4.1
 - Tokenizers: 0.22.1
 

adapter_config.json

@@ -16,7 +16,7 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 256,
+  "lora_alpha": 16,
   "lora_bias": false,
   "lora_dropout": 0.05,
   "megatron_config": null,
@@ -25,19 +25,12 @@
   "peft_type": "LORA",
   "peft_version": "0.18.0",
   "qalora_group_size": 16,
-  "r": 128,
+  "r": 256,
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "embed_tokens",
-    "k_proj",
-    "q_proj",
-    "o_proj",
-    "gate_proj",
-    "down_proj",
-    "lm_head",
-    "up_proj",
-    "v_proj"
+    "all-linear",
+    "embed_tokens"
   ],
   "target_parameters": null,
   "task_type": "CAUSAL_LM",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:af00212be9a7dcf50cf91855d57bed865d94c4a151252e80042aa013da256c77
-size 6489752072
+oid sha256:d6db50ce8797893a87b0bff0a640b7bdef6d557a5eddaa1341d3a435a4fa031a
+size 2285694936

chat_template.jinja

@@ -2,11 +2,9 @@
 {%- if messages[0]['role'] == 'system' -%}
     {%- if messages[0]['content'] is string -%}
         {%- set first_user_prefix = messages[0]['content'] + '
-
 ' -%}
     {%- else -%}
         {%- set first_user_prefix = messages[0]['content'][0]['text'] + '
-
 ' -%}
     {%- endif -%}
     {%- set loop_messages = messages[1:] -%}
@@ -15,21 +13,30 @@
     {%- set loop_messages = messages -%}
 {%- endif -%}
 {%- for message in loop_messages -%}
-    {%- if (message['role'] == 'assistant') -%}
+    {%- if message['role'] == 'assistant' or message['role'] == 'model' -%}
         {%- set role = "model" -%}
     {%- else -%}
         {%- set role = message['role'] -%}
     {%- endif -%}
+
     {{ '<start_of_turn>' + role + '
 ' + (first_user_prefix if loop.first else "") }}
     {%- if message['content'] is string -%}
-        {{ message['content'] | trim }}
+        {%- if role == "model" -%}
+            {% generation %}    {{ message['content'] | trim }}     {% endgeneration %}
+        {%- else -%}
+            {{ message['content'] | trim }}
+        {%- endif -%}
     {%- elif message['content'] is iterable -%}
         {%- for item in message['content'] -%}
             {%- if item['type'] == 'image' -%}
                 {{ '<start_of_image>' }}
             {%- elif item['type'] == 'text' -%}
-                {{ item['text'] | trim }}
+                {%- if role == "model" -%}
+                    {% generation %}    {{ item['text'] | trim }}   {% endgeneration %}
+                {%- else -%}
+                    {{ item['text'] | trim }}
+                {%- endif -%}
             {%- endif -%}
         {%- endfor -%}
     {%- else -%}

special_tokens_map.json

@@ -16,13 +16,7 @@
     "single_word": false
   },
   "eoi_token": "<end_of_image>",
-  "eos_token": {
-    "content": "<eos>",
-    "lstrip": false,
-    "normalized": false,
-    "rstrip": false,
-    "single_word": false
-  },
+  "eos_token": "<eos>",
   "image_token": "<image_soft_token>",
   "pad_token": {
     "content": "<pad>",

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:c6ab89697ec2b7f984bc8f0b5136321c8dd877057cf0372cd31ec1aa32e9b5f6
+oid sha256:0ff25ad6383f18b7fe891c349f04b4502488f62babf7cd9551544ff0e6b3f6da
 size 6225