From 7ac74e6d5810d02c6d0e3fef0e1f86f9e7d2b217 Mon Sep 17 00:00:00 2001
From: Rujikorn Charakorn <rujikorn.ch@gmail.com>
Date: Tue, 26 Aug 2025 18:14:40 +0900
Subject: [PATCH] SFT data (#13)

* sft data from smoltalk + multi-turn ctx data

* add ctx-q-gsm (stateful-gsm from sleep-time compute)
---
 src/ctx_to_lora/data/definitions.py      | 19 ++++++++++++
 src/ctx_to_lora/data/preprocessing_fn.py | 37 ++++++++++++++++++++++++
 src/ctx_to_lora/data/processing.py       |  6 ++--
 3 files changed, 60 insertions(+), 2 deletions(-)

diff --git a/src/ctx_to_lora/data/definitions.py b/src/ctx_to_lora/data/definitions.py
index 0b911fe..02b6f1f 100644
--- a/src/ctx_to_lora/data/definitions.py
+++ b/src/ctx_to_lora/data/definitions.py
@@ -581,6 +581,25 @@ DS_KWARGS = {
             split="train",
         ),
     ),
+    # sft data
+    "openhermes": dict(
+        train=dict(path="HuggingFaceTB/smoltalk", name="openhermes-100k", split="train")
+    ),
+    "smol-magpie-ultra": dict(
+        train=dict(
+            path="HuggingFaceTB/smoltalk", name="smol-magpie-ultra", split="train"
+        )
+    ),
+    "smol-rewrite": dict(
+        train=dict(path="HuggingFaceTB/smoltalk", name="smol-rewrite", split="train")
+    ),
+    "smol-summarize": dict(
+        train=dict(path="HuggingFaceTB/smoltalk", name="smol-summarize", split="train")
+    ),
+    "systemchats": dict(
+        train=dict(path="HuggingFaceTB/smoltalk", name="systemchats-30k", split="train")
+    ),
+    "ctx-q-gsm": dict(train=dict(path="letta-ai/stateful-gsm-symbolic", split="train")),
 }
 
 # add ctx_numbers
diff --git a/src/ctx_to_lora/data/preprocessing_fn.py b/src/ctx_to_lora/data/preprocessing_fn.py
index 1c61f15..fd81a61 100644
--- a/src/ctx_to_lora/data/preprocessing_fn.py
+++ b/src/ctx_to_lora/data/preprocessing_fn.py
@@ -250,6 +250,43 @@ def get_preprocessing_fn(
                 "prompt": sample["instruction"],
                 "response": "```python\n" + sample["code"].strip() + "\n```",
             }
+    elif "openhermes" == ds_name:
+        # system prompt is from a set of predefined prompts → query
+        # user prompt is the content → ctx
+        def f(sample):
+            messages = sample["messages"]
+            if messages[0]["role"] != "system":
+                return {"context": None, "prompt": None, "response": None}
+            ctx = messages[1]["content"]  # user input
+            q = messages[0]["content"]  # system intx
+            response = messages[2]["content"]
+            return {"context": ctx, "prompt": q, "response": response}
+
+    elif "smol-magpie-ultra" == ds_name or "systemchats" == ds_name:
+        # ctx → all history up to the last turn
+        # q → last user input
+        def f(sample):
+            chat_history = sample["messages"][:-2]
+            q = sample["messages"][-2]["content"]
+            response = sample["messages"][-1]["content"]
+            return {"context": chat_history, "prompt": q, "response": response}
+
+    elif "smol-rewrite" == ds_name or "smol-summarize" == ds_name:
+        # user content is a paragraph of text → ctx
+        # system prompt is a paraphrase/summarize instruction → q
+        def f(sample):
+            ctx = sample["messages"][1]["content"]
+            q = sample["messages"][0]["content"]
+            response = sample["messages"][2]["content"]
+            return {"context": ctx, "prompt": q, "response": response}
+
+    elif "ctx-q-gsm" == ds_name:
+
+        def f(sample):
+            ctx = sample["stateful_gsm_symbolic_context"]
+            q = sample["stateful_gsm_symbolic_question"]
+            response = sample["answer"]
+            return {"context": ctx, "prompt": q, "response": response}
 
     if is_eval and (ds_name in EVAL_INTX_TEMPLATES):
         prompt_template = EVAL_INTX_TEMPLATES[ds_name]
diff --git a/src/ctx_to_lora/data/processing.py b/src/ctx_to_lora/data/processing.py
index ba58dcf..d892efa 100644
--- a/src/ctx_to_lora/data/processing.py
+++ b/src/ctx_to_lora/data/processing.py
@@ -873,9 +873,11 @@ def tokenize_ctx_text(
             [
                 [
                     {"role": "system", "content": ""},
-                    {"role": "user", "content": text.strip()},
+                    {"role": "user", "content": ctx.strip()},
                 ]
-                for text in samples["context"]
+                if isinstance(ctx, str)
+                else ctx
+                for ctx in samples["context"]
             ],
             tokenize=True,
             add_generation_prompt=True,