Training in progress, step 57

README.md

@@ -1,25 +1,17 @@
 ---
 base_model: openai/gpt-oss-20b
-datasets:
-- kingabzpro/dermatology-qa-firecrawl-dataset
 library_name: transformers
 model_name: gpt-oss-20b-dermatology-qa
 tags:
 - generated_from_trainer
-- trl
 - sft
-- dermatology
-- medical
+- trl
 licence: license
-license: apache-2.0
-language:
-- en
-pipeline_tag: text-generation
 ---
 
 # Model Card for gpt-oss-20b-dermatology-qa
 
-This model is a fine-tuned version of [openai/gpt-oss-20b](https://huggingface.co/openai/gpt-oss-20b) on the [kingabzpro/dermatology-qa-firecrawl-dataset](https://huggingface.co/kingabzpro/gpt-oss-20b-medical-qa) dataset.
+This model is a fine-tuned version of [openai/gpt-oss-20b](https://huggingface.co/openai/gpt-oss-20b).
 It has been trained using [TRL](https://github.com/huggingface/trl).
 
 ## Quick start
@@ -27,25 +19,40 @@ It has been trained using [TRL](https://github.com/huggingface/trl).
 ```python
 from transformers import pipeline
 
-question = "How does the source suggest clinicians approach the diagnosis of rosacea?"
+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="kingabzpro/gpt-oss-20b-dermatology-qa", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
+```
 
-# Load pipeline
-generator = pipeline(
-    "text-generation",
-    model="kingabzpro/gpt-oss-20b-dermatology-qa",
-    device="cuda"  # or device=0
-)
+## Training procedure
 
+ 
 
-# Run inference (passing in chat-style format)
-output = generator(
-    [{"role": "user", "content": question}],
-    max_new_tokens=200,
-    return_full_text=False
-)[0]
 
-print(output["generated_text"])
+This model was trained with SFT.
+
+### Framework versions
+
+- TRL: 0.22.1
+- Transformers: 4.55.4
+- Pytorch: 2.8.0.dev20250319+cu128
+- Datasets: 4.0.0
+- Tokenizers: 0.21.4
+
+## Citations
+
 
-# The source says that clinicians should use a combination of clinical signs and symptoms when diagnosing rosacea, rather than relying on a single feature.
-```
 
+Cite TRL as:
+    
+```bibtex
+@misc{vonwerra2022trl,
+	title        = {{TRL: Transformer Reinforcement Learning}},
+	author       = {Leandro von Werra and Younes Belkada and Lewis Tunstall and Edward Beeching and Tristan Thrush and Nathan Lambert and Shengyi Huang and Kashif Rasul and Quentin Gallou{\'e}dec},
+	year         = 2020,
+	journal      = {GitHub repository},
+	publisher    = {GitHub},
+	howpublished = {\url{https://github.com/huggingface/trl}}
+}
+```

adapter_config.json

@@ -28,10 +28,10 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "k_proj",
     "q_proj",
-    "v_proj",
     "o_proj",
-    "k_proj"
+    "v_proj"
   ],
   "target_parameters": [
     "7.mlp.experts.gate_up_proj",

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8e1cef7449b0e192070e9f93ae7ac50e03ff6bdfb228429bdcad06a12e9203fb
+oid sha256:b0d9138da1714e204f2ad4ebcc8d8519ab4e85b7096299a12bd96c8b34f64abd
 size 60189176

runs/Sep03_09-59-57_61d8ce0bbd73/events.out.tfevents.1756893598.61d8ce0bbd73.321.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e7a8af51b682bc0b8f8bea3881f863eaf40b4169e2df14b9086e0b5e8b9b7d88
+size 6494

runs/Sep03_10-01-15_61d8ce0bbd73/events.out.tfevents.1756893678.61d8ce0bbd73.321.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:66120d3af39f438c59f7488b67ef4fbab0257a800b6298a56fb37638a966c46d
+size 6493

runs/Sep03_10-02-29_61d8ce0bbd73/events.out.tfevents.1756893751.61d8ce0bbd73.861.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6e38fa4ebfae282e00c2ecd8c611b90fa643204a08cb197936cd65adfa49d30e
+size 6493

runs/Sep03_10-03-20_61d8ce0bbd73/events.out.tfevents.1756893802.61d8ce0bbd73.861.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd2441a752b1cca01b3b12f09207151a79d40b693d6427333ffda9961776274f
+size 6494

runs/Sep03_10-04-01_61d8ce0bbd73/events.out.tfevents.1756893843.61d8ce0bbd73.1127.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:623c3b00a8b13436073d67bd6559891f69da345711326322ddb6075c580ec3d9
+size 6494

runs/Sep03_10-11-55_61d8ce0bbd73/events.out.tfevents.1756894418.61d8ce0bbd73.1387.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3ae3c924043441369e0d32654be986e5b14e355880d5f4435b0fe46b60e246da
+size 6494

runs/Sep03_10-11-55_61d8ce0bbd73/events.out.tfevents.1756894532.61d8ce0bbd73.1387.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c516e8a1b32925c2c7e19a4d7e5470661f8df8c761d85251e9f53ea979985d45
+size 6494

runs/Sep03_10-17-05_61d8ce0bbd73/events.out.tfevents.1756894627.61d8ce0bbd73.1856.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3697b0640edc4b0e05c48a1de044de9edc5dd57f8016a8a3ad8dd34038fb2f7e
+size 6494

runs/Sep03_10-17-54_61d8ce0bbd73/events.out.tfevents.1756894678.61d8ce0bbd73.1856.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:760be4a96178a0030d827cdb852536210681bbf45eb8f7f774254e8e510e082d
+size 6494

runs/Sep03_10-18-45_61d8ce0bbd73/events.out.tfevents.1756894727.61d8ce0bbd73.2432.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7c0dd2027b1de5dd789edc1b7c50dea9155a3b2f8edbb19062cd598d395ec85e
+size 6494

runs/Sep03_10-26-32_61d8ce0bbd73/events.out.tfevents.1756895193.61d8ce0bbd73.2432.1

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a43e4197583a728237c866d45b02cfa5a2057229f7bcf2910b0341f589d0463a
+size 6493

runs/Sep03_10-33-17_61d8ce0bbd73/events.out.tfevents.1756895598.61d8ce0bbd73.3117.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:47574b3c6522c1505cdcb32e66b7a982bd5aa55174761c078b2e2a61eb5c3a39
+size 10990

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9a6cd323fa2dfe2d16f9d2d0825be8b36bbafe8b49082603f84b3b49d0675d83
+oid sha256:11adc9de4fb4f041b65e1f4c7ce061b9c334f015eb48dc8e24888faf14e0d75b
 size 6161