End of training

README.md

@@ -5,11 +5,11 @@ library_name: transformers
 model_name: trainer_output
 tags:
 - generated_from_trainer
-- tool-use
 - sft
+- tool-use
 - unsloth
-- linalg-zero
 - trl
+- linalg-zero
 licence: license
 ---
 

adapter_config.json

@@ -29,13 +29,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "v_proj",
+    "q_proj",
+    "k_proj",
     "gate_proj",
     "o_proj",
     "down_proj",
-    "k_proj",
-    "q_proj",
-    "up_proj",
-    "v_proj"
+    "up_proj"
   ],
   "task_type": "CAUSAL_LM",
   "trainable_token_indices": null,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:b0156b8e2bd7ee00762484b1b56e008189907477eea7e35d71fabc479be41072
+oid sha256:627e6582769658677afc0094ca2ab8e64ae71587ad8d11d9168e04e0be8f907c
 size 132188392

all_results.json

@@ -1,14 +1,14 @@
 {
     "epoch": 1.0,
-    "eval_loss": 0.5140631794929504,
-    "eval_runtime": 8.3035,
+    "eval_loss": 0.8699710965156555,
+    "eval_runtime": 8.6196,
     "eval_samples": 60,
-    "eval_samples_per_second": 7.226,
-    "eval_steps_per_second": 7.226,
+    "eval_samples_per_second": 6.961,
+    "eval_steps_per_second": 6.961,
     "total_flos": 2717720554039296.0,
-    "train_loss": 0.24174172580242156,
-    "train_runtime": 31.6344,
+    "train_loss": 0.0,
+    "train_runtime": 0.0042,
     "train_samples": 60,
-    "train_samples_per_second": 1.897,
-    "train_steps_per_second": 1.897
+    "train_samples_per_second": 14376.363,
+    "train_steps_per_second": 239.606
 }

chat_template.jinja

@@ -1 +1,85 @@
-{% if messages[0]['role'] == 'system' %}{{ messages[0]['content'] + eos_token }}{% set loop_messages = messages[1:] %}{% else %}{{ 'You are a helpful assistant.' + eos_token }}{% set loop_messages = messages %}{% endif %}{% for message in loop_messages %}{% if message['role'] == 'user' %}{{ message['content'] }}{% elif message['role'] == 'assistant' %}{{ message['content'] + eos_token }}{% endif %}{% endfor %}{% if add_generation_prompt %}{{ '<start_working_out>' }}{% endif %}
+{%- if tools %}
+    {{- '<|im_start|>system\n' }}
+    {%- if messages[0].role == 'system' %}
+        {{- messages[0].content + '\n\n' }}
+    {%- endif %}
+    {{- "# Tools\n\nYou may call one or more functions to assist with the user query.\n\nYou are provided with function signatures within <tools></tools> XML tags:\n<tools>" }}
+    {%- for tool in tools %}
+        {{- "\n" }}
+        {{- tool | tojson }}
+    {%- endfor %}
+    {{- "\n</tools>\n\nFor each function call, return a json object with function name and arguments within <tool_call></tool_call> XML tags:\n<tool_call>\n{\"name\": <function-name>, \"arguments\": <args-json-object>}\n</tool_call><|im_end|>\n" }}
+{%- else %}
+    {%- if messages[0].role == 'system' %}
+        {{- '<|im_start|>system\n' + messages[0].content + '<|im_end|>\n' }}
+    {%- endif %}
+{%- endif %}
+{%- set ns = namespace(multi_step_tool=true, last_query_index=messages|length - 1) %}
+{%- for message in messages[::-1] %}
+    {%- set index = (messages|length - 1) - loop.index0 %}
+    {%- if ns.multi_step_tool and message.role == "user" and not(message.content.startswith('<tool_response>') and message.content.endswith('</tool_response>')) %}
+        {%- set ns.multi_step_tool = false %}
+        {%- set ns.last_query_index = index %}
+    {%- endif %}
+{%- endfor %}
+{%- for message in messages %}
+    {%- if (message.role == "user") or (message.role == "system" and not loop.first) %}
+        {{- '<|im_start|>' + message.role + '\n' + message.content + '<|im_end|>' + '\n' }}
+    {%- elif message.role == "assistant" %}
+        {%- set content = message.content %}
+        {%- set reasoning_content = '' %}
+        {%- if message.reasoning_content is defined and message.reasoning_content is not none %}
+            {%- set reasoning_content = message.reasoning_content %}
+        {%- else %}
+            {%- if '</think>' in message.content %}
+                {%- set content = message.content.split('</think>')[-1].lstrip('\n') %}
+                {%- set reasoning_content = message.content.split('</think>')[0].rstrip('\n').split('<think>')[-1].lstrip('\n') %}
+            {%- endif %}
+        {%- endif %}
+        {%- if loop.index0 > ns.last_query_index %}
+            {%- if loop.last or (not loop.last and reasoning_content) %}
+                {{- '<|im_start|>' + message.role + '\n<think>\n' + reasoning_content.strip('\n') + '\n</think>\n\n' + content.lstrip('\n') }}
+            {%- else %}
+                {{- '<|im_start|>' + message.role + '\n' + content }}
+            {%- endif %}
+        {%- else %}
+            {{- '<|im_start|>' + message.role + '\n' + content }}
+        {%- endif %}
+        {%- if message.tool_calls %}
+            {%- for tool_call in message.tool_calls %}
+                {%- if (loop.first and content) or (not loop.first) %}
+                    {{- '\n' }}
+                {%- endif %}
+                {%- if tool_call.function %}
+                    {%- set tool_call = tool_call.function %}
+                {%- endif %}
+                {{- '<tool_call>\n{"name": "' }}
+                {{- tool_call.name }}
+                {{- '", "arguments": ' }}
+                {%- if tool_call.arguments is string %}
+                    {{- tool_call.arguments }}
+                {%- else %}
+                    {{- tool_call.arguments | tojson }}
+                {%- endif %}
+                {{- '}\n</tool_call>' }}
+            {%- endfor %}
+        {%- endif %}
+        {{- '<|im_end|>\n' }}
+    {%- elif message.role == "tool" %}
+        {%- if loop.first or (messages[loop.index0 - 1].role != "tool") %}
+            {{- '<|im_start|>user' }}
+        {%- endif %}
+        {{- '\n<tool_response>\n' }}
+        {{- message.content }}
+        {{- '\n</tool_response>' }}
+        {%- if loop.last or (messages[loop.index0 + 1].role != "tool") %}
+            {{- '<|im_end|>\n' }}
+        {%- endif %}
+    {%- endif %}
+{%- endfor %}
+{%- if add_generation_prompt %}
+    {{- '<|im_start|>assistant\n' }}
+    {%- if enable_thinking is defined and enable_thinking is false %}
+        {{- '<think>\n\n</think>\n\n' }}
+    {%- endif %}
+{%- endif %}

eval_results.json

@@ -1,8 +1,8 @@
 {
     "epoch": 1.0,
-    "eval_loss": 0.5140631794929504,
-    "eval_runtime": 8.3035,
+    "eval_loss": 0.8699710965156555,
+    "eval_runtime": 8.6196,
     "eval_samples": 60,
-    "eval_samples_per_second": 7.226,
-    "eval_steps_per_second": 7.226
+    "eval_samples_per_second": 6.961,
+    "eval_steps_per_second": 6.961
 }

train_results.json

@@ -1,9 +1,9 @@
 {
     "epoch": 1.0,
     "total_flos": 2717720554039296.0,
-    "train_loss": 0.24174172580242156,
-    "train_runtime": 31.6344,
+    "train_loss": 0.0,
+    "train_runtime": 0.0042,
     "train_samples": 60,
-    "train_samples_per_second": 1.897,
-    "train_steps_per_second": 1.897
+    "train_samples_per_second": 14376.363,
+    "train_steps_per_second": 239.606
 }

trainer_state.json

@@ -3,111 +3,35 @@
   "best_metric": null,
   "best_model_checkpoint": null,
   "epoch": 1.0,
-  "eval_steps": 500,
-  "global_step": 60,
+  "eval_steps": 50,
+  "global_step": 1,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 0.08333333333333333,
-      "grad_norm": 0.3362419009208679,
-      "learning_rate": 0.00016,
-      "loss": 0.7526,
-      "step": 5
-    },
-    {
-      "epoch": 0.16666666666666666,
-      "grad_norm": 0.5312803983688354,
-      "learning_rate": 0.00018545454545454545,
-      "loss": 0.4235,
-      "step": 10
-    },
-    {
-      "epoch": 0.25,
-      "grad_norm": 0.3434021770954132,
-      "learning_rate": 0.00016727272727272728,
-      "loss": 0.2288,
-      "step": 15
-    },
-    {
-      "epoch": 0.3333333333333333,
-      "grad_norm": 0.26192706823349,
-      "learning_rate": 0.0001490909090909091,
-      "loss": 0.1897,
-      "step": 20
-    },
-    {
-      "epoch": 0.4166666666666667,
-      "grad_norm": 0.26772984862327576,
-      "learning_rate": 0.00013090909090909093,
-      "loss": 0.1833,
-      "step": 25
-    },
-    {
-      "epoch": 0.5,
-      "grad_norm": 0.20748060941696167,
-      "learning_rate": 0.00011272727272727272,
-      "loss": 0.1621,
-      "step": 30
-    },
-    {
-      "epoch": 0.5833333333333334,
-      "grad_norm": 0.21864831447601318,
-      "learning_rate": 9.454545454545455e-05,
-      "loss": 0.1638,
-      "step": 35
-    },
-    {
-      "epoch": 0.6666666666666666,
-      "grad_norm": 0.24830199778079987,
-      "learning_rate": 7.636363636363637e-05,
-      "loss": 0.1667,
-      "step": 40
-    },
-    {
-      "epoch": 0.75,
-      "grad_norm": 0.20934940874576569,
-      "learning_rate": 5.818181818181818e-05,
-      "loss": 0.1678,
-      "step": 45
-    },
-    {
-      "epoch": 0.8333333333333334,
-      "grad_norm": 0.2636692523956299,
-      "learning_rate": 4e-05,
-      "loss": 0.1548,
-      "step": 50
-    },
-    {
-      "epoch": 0.9166666666666666,
-      "grad_norm": 0.1722150295972824,
-      "learning_rate": 2.1818181818181818e-05,
-      "loss": 0.1524,
-      "step": 55
-    },
-    {
-      "epoch": 1.0,
-      "grad_norm": 0.23989179730415344,
-      "learning_rate": 3.636363636363636e-06,
-      "loss": 0.1553,
-      "step": 60
+      "epoch": 0,
+      "eval_loss": 0.8395363092422485,
+      "eval_runtime": 8.5871,
+      "eval_samples_per_second": 6.987,
+      "eval_steps_per_second": 6.987,
+      "step": 0
     },
     {
       "epoch": 1.0,
-      "step": 60,
+      "step": 1,
       "total_flos": 2717720554039296.0,
-      "train_loss": 0.24174172580242156,
-      "train_runtime": 31.6344,
-      "train_samples_per_second": 1.897,
-      "train_steps_per_second": 1.897
+      "train_loss": 0.0,
+      "train_runtime": 0.0042,
+      "train_samples_per_second": 14376.363,
+      "train_steps_per_second": 239.606
     }
   ],
   "logging_steps": 5,
-  "max_steps": 60,
+  "max_steps": 1,
   "num_input_tokens_seen": 0,
   "num_train_epochs": 1,
-  "save_steps": 500,
+  "save_steps": 50,
   "stateful_callbacks": {
     "TrainerControl": {
       "args": {

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:54b494cceb94d5932b6c823a353a9515458fd050c35e966606a622ff2776e061
+oid sha256:0cff4ee75012ee385e734726f3dee6d7d4a60147422cc2344287ce24a45db845
 size 6097