tpircsc/phi-4-mini-it-thinking-function_calling-V0

README.md

@@ -1,4 +1,5 @@
 ---
+base_model: microsoft/Phi-4-mini-instruct
 library_name: transformers
 model_name: phi-4-mini-it-thinking-function_calling-V0
 tags:
@@ -6,17 +7,11 @@ tags:
 - trl
 - sft
 licence: license
-datasets:
-- Jofthomas/hermes-function-calling-thinking-V1
-pipeline_tag: text-generation
 ---
 
 # Model Card for phi-4-mini-it-thinking-function_calling-V0
 
 This model is a fine-tuned version of [microsoft/Phi-4-mini-instruct](https://huggingface.co/microsoft/Phi-4-mini-instruct).
-
-It uses a modified subset of the [NousResearch/hermes-function-calling-v1](https://huggingface.co/datasets/NousResearch/hermes-function-calling-v1) dataset: [Jofthomas/hermes-function-calling-thinking-V1](https://huggingface.co/datasets/Jofthomas/hermes-function-calling-thinking-V1), which adds a thinking step to the base dataset.
-
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start

adapter_config.json

@@ -12,7 +12,7 @@
   "layers_pattern": null,
   "layers_to_transform": null,
   "loftq_config": {},
-  "lora_alpha": 64,
+  "lora_alpha": 32,
   "lora_bias": false,
   "lora_dropout": 0.05,
   "megatron_config": null,
@@ -23,15 +23,10 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
-    "gate_proj",
-    "k_proj",
-    "down_proj",
-    "up_proj",
+    "qkv_proj",
     "o_proj",
-    "v_proj",
-    "embed_tokens",
-    "q_proj",
-    "lm_head"
+    "down_proj",
+    "gate_up_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dfc16a3056ec1f0655fddd8da6baeabe5da3a9c5ee86b1626661573ee7d52858
-size 2519746832
+oid sha256:a620777428885e25225eccc07d8caf21da2d08c3d0d6d4d7c4b11b4898920a86
+size 2550290272

added_tokens.json

@@ -1,14 +1,6 @@
 {
-  "</think>": 200033,
-  "</tool_call>": 200035,
-  "</tool_reponse>": 200037,
-  "</tools>": 200031,
-  "<eos>": 200038,
-  "<pad>": 200029,
-  "<think>": 200032,
-  "<tool_call>": 200034,
-  "<tool_reponse>": 200036,
-  "<tools>": 200030,
+  "</think>": 200030,
+  "<think>": 200029,
   "<|/tool_call|>": 200026,
   "<|/tool|>": 200024,
   "<|assistant|>": 200019,

runs/Mar11_17-21-08_73fdee6cedcf/events.out.tfevents.1741713685.73fdee6cedcf.803.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:633e04ac6a416d8a39617ab78df3ce17b1c0d167072f083a09b796cdad1158ff
+size 11917

special_tokens_map.json

@@ -1,15 +1,7 @@
 {
   "additional_special_tokens": [
-    "<tools>",
-    "</tools>",
     "<think>",
-    "</think>",
-    "<tool_call>",
-    "</tool_call>",
-    "<tool_reponse>",
-    "</tool_reponse>",
-    "<pad>",
-    "<eos>"
+    "</think>"
   ],
   "bos_token": {
     "content": "<|endoftext|>",
@@ -18,8 +10,20 @@
     "rstrip": false,
     "single_word": false
   },
-  "eos_token": "<eos>",
-  "pad_token": "<pad>",
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
   "unk_token": {
     "content": "<|endoftext|>",
     "lstrip": false,

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b3ba28fcbeb095731859469f3fb1419742a2fea50d727f5352ffc4505c0e6dde
-size 15525957
+oid sha256:c5c2a1454e12b996fb8e3def8c3770f1ab45e4a38fede20471d0e557e6de8dc7
+size 15524464

tokenizer_config.json

@@ -100,30 +100,6 @@
       "special": true
     },
     "200029": {
-      "content": "<pad>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "200030": {
-      "content": "<tools>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "200031": {
-      "content": "</tools>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "200032": {
       "content": "<think>",
       "lstrip": false,
       "normalized": false,
@@ -131,74 +107,26 @@
       "single_word": false,
       "special": true
     },
-    "200033": {
+    "200030": {
       "content": "</think>",
       "lstrip": false,
       "normalized": false,
       "rstrip": false,
       "single_word": false,
       "special": true
-    },
-    "200034": {
-      "content": "<tool_call>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "200035": {
-      "content": "</tool_call>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "200036": {
-      "content": "<tool_reponse>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "200037": {
-      "content": "</tool_reponse>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "200038": {
-      "content": "<eos>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
     }
   },
   "additional_special_tokens": [
-    "<tools>",
-    "</tools>",
     "<think>",
-    "</think>",
-    "<tool_call>",
-    "</tool_call>",
-    "<tool_reponse>",
-    "</tool_reponse>",
-    "<pad>",
-    "<eos>"
+    "</think>"
   ],
   "bos_token": "<|endoftext|>",
-  "chat_template": "{{ bos_token }}{% if messages[0]['role'] == 'system' %}{{ raise_exception('System role not supported') }}{% endif %}{% for message in messages %}{{ '<start_of_turn>' + message['role'] + '\n' + message['content'] | trim + '<end_of_turn><eos>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}",
+  "chat_template": "{% for message in messages %}{% if message['role'] == 'system' and 'tools' in message and message['tools'] is not none %}{{ '<|' + message['role'] + '|>' + message['content'] + '<|tool|>' + message['tools'] + '<|/tool|>' + '<|end|>' }}{% else %}{{ '<|' + message['role'] + '|>' + message['content'] + '<|end|>' }}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '<|assistant|>' }}{% else %}{{ eos_token }}{% endif %}",
   "clean_up_tokenization_spaces": false,
-  "eos_token": "<eos>",
+  "eos_token": "<|endoftext|>",
   "extra_special_tokens": {},
   "model_max_length": 131072,
-  "pad_token": "<pad>",
+  "pad_token": "<|endoftext|>",
   "tokenizer_class": "GPT2Tokenizer",
   "unk_token": "<|endoftext|>"
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f88d8e1e94b68ee14fd1ac639f1e3c2d9831ab942483f1651b812349959da896
+oid sha256:c887b52524c824bc3b3257e60c6b44085b290550950629322dc7b3a7d12dc069
 size 5624