Upload folder using huggingface_hub

README.md

@@ -1,133 +1,58 @@
 ---
-library_name: transformers
-license: apache-2.0
 base_model: openai/gpt-oss-20b
+library_name: transformers
+model_name: gpt-oss-20b-multilingual-reasoner
 tags:
 - generated_from_trainer
-datasets:
-- HuggingFaceH4/Multilingual-Thinking
-model-index:
-- name: workspace/data/outputs/gpt-oss-out/
-  results: []
+- trl
+- sft
+licence: license
 ---
 
-<!-- This model card has been generated automatically according to the information the Trainer had access to. You
-should probably proofread and complete it, then remove this comment. -->
-
-[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
-<details><summary>See axolotl config</summary>
-
-axolotl version: `0.12.0.dev0`
-```yaml
-base_model: openai/gpt-oss-20b
-use_kernels: true
-model_quantization_config: Mxfp4Config
-model_quantization_config_kwargs:
-  dequantize: true
-
-plugins:
-  - axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin
-
-experimental_skip_move_to_device: true  # prevent OOM by NOT putting model to GPU before sharding
-
-datasets:
-  - path: HuggingFaceH4/Multilingual-Thinking
-    type: chat_template
-    field_thinking: thinking
-    template_thinking_key: thinking
-
-dataset_prepared_path: last_run_prepared
-val_set_size: 0
-output_dir: /workspace/data/outputs/gpt-oss-out/
-
-sequence_len: 8196
-sample_packing: true
-pad_to_sequence_len: true
-
-wandb_project: gpt-oss-20b
-wandb_name: multilingual-reasoning-fft
-
-gradient_accumulation_steps: 1
-micro_batch_size: 2
-num_epochs: 1
-
-optimizer: adamw_torch_fused
-lr_scheduler: constant_with_warmup
-learning_rate: 2e-5
-
-bf16: true
-tf32: true
+# Model Card for gpt-oss-20b-multilingual-reasoner
 
-flash_attention: true
-attn_implementation: kernels-community/vllm-flash-attn3
+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).
 
-gradient_checkpointing: true
-#activation_offloading: true
+## Quick start
 
-logging_steps: 1
-saves_per_epoch: 1
+```python
+from transformers import pipeline
 
-warmup_ratio: 0.03
-
-special_tokens:
-eot_tokens:
-  - "<|end|>"
-  - "<|return|>"
-
-deepspeed: /pi-workspace/zero3.json
-
-# fsdp_version: 2
-# fsdp_config:
-#   offload_params: false
-#   state_dict_type: SHARDED_STATE_DICT
-#   auto_wrap_policy: TRANSFORMER_BASED_WRAP
-#   transformer_layer_cls_to_wrap: GptOssDecoderLayer
-#   reshard_after_forward: true
-# #  cpu_ram_efficient_loading: true
+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="None", device="cuda")
+output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
+print(output["generated_text"])
 ```
 
-</details><br>
-
-# workspace/data/outputs/gpt-oss-out/
-
-This model is a fine-tuned version of [openai/gpt-oss-20b](https://huggingface.co/openai/gpt-oss-20b) on the HuggingFaceH4/Multilingual-Thinking dataset.
-
-## Model description
-
-More information needed
-
-## Intended uses & limitations
-
-More information needed
+## Training procedure
 
-## Training and evaluation data
+[<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="150" height="24"/>](https://wandb.ai/b37h3z3n/huggingface/runs/4059q72j) 
 
-More information needed
 
-## Training procedure
+This model was trained with SFT.
 
-### Training hyperparameters
-
-The following hyperparameters were used during training:
-- learning_rate: 2e-05
-- train_batch_size: 2
-- eval_batch_size: 2
-- seed: 42
-- distributed_type: multi-GPU
-- num_devices: 8
-- total_train_batch_size: 16
-- total_eval_batch_size: 16
-- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
-- lr_scheduler_type: constant_with_warmup
-- training_steps: 8
+### Framework versions
 
-### Training results
+- TRL: 0.21.0
+- Transformers: 4.55.0
+- Pytorch: 2.8.0+cu128
+- Datasets: 4.0.0
+- Tokenizers: 0.21.4
 
+## Citations
 
 
-### Framework versions
 
-- Transformers 4.55.0
-- Pytorch 2.8.0+cu128
-- Datasets 4.0.0
-- Tokenizers 0.21.4
+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}}
+}
+```

model-00001-of-00009.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:1e66b52040a9a0777a9c94ad800d7dff662a58c96129efbfd77b54ef553560bd
+oid sha256:c1ef9a81309d602c303a1cb43b568d9e4a2faf761cb676b62af26182511e1452
 size 4504304664

model-00002-of-00009.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9cbe12bb05ec8772ff87808f8916a52f9c330075813be3a978e6b968ba2aa52b
+oid sha256:56dd70d969c9f77ce0b7e79424919d09db0882cb22a2ea7fcd16a7c7267e6530
 size 4939127656

model-00003-of-00009.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:77057a058259fdad6458a2cc253c248a9a308c43390b2e1eaeff108144d1502e
+oid sha256:a535b6916849716c429e46d9e8c190ba7c49e54b48bd070416fb42ff3e0c3129
 size 4939127656

model-00004-of-00009.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:72cc1b6291eec305ae8b01a4f7ab53eb15489496d4fe36b7613d822cf307ccf2
+oid sha256:ec38b8349e47cee4a7fc123845e10dab215b123ed15a7090465c93558ebcf158
 size 4939127680

model-00005-of-00009.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:7c0775267ff51f292d952a8a06d7d7a4cb709ea9fa48433f5dde5dc1ead0d84f
+oid sha256:ac94173d30a5de5a17382c79bf0cabef29c1d34d27fba3128e701ba9aa012e4c
 size 4939127704

model-00006-of-00009.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:89e821b90576a93e94a9457d21012a184e3d3336cd8a499b5040dcc9fa8791c8
+oid sha256:0ccc09d6d016bef393bfb1f024a448a9a5b51db9d15dec718ba3af819288efb1
 size 4939127704

model-00007-of-00009.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:3f2b0083866777b6ef7405acca40a46242005bbebb423f09bd3223880cc94ee8
+oid sha256:3c9ca478c80cd187756bae63154e5afec330448bd59c6969937bbcba8ad2ed7b
 size 4939127704

model-00008-of-00009.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2b97d39f740e42cf0bbe9e15a6ad59dae1b93396ca2d85899c4bb9d0b6882192
+oid sha256:069f7d27321f545ede37d5716ce4d2dd1aa6f21728a81a9c5a499a76d789d893
 size 4939127704

model-00009-of-00009.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:549eb052e40666b8b016c7a6db7031e752785cec977beed6a315fea12489caef
+oid sha256:4d0722a3185a07dabe2c008116e476336449ba7606dc64e7cc8a4435c508c7e5
 size 2751362856

model.safetensors.index.json

@@ -1,6 +1,6 @@
 {
   "metadata": {
-    "total_parameters": 335424,
+    "total_parameters": 4759104,
     "total_size": 41829514368
   },
   "weight_map": {

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:29e1c313582c8dd3b40fa45dcf6d6482aeabf058adc5837643ba6a5b2ecdb37c
-size 9489
+oid sha256:5a98f69af00540d86783a5d39f060a49f94d8d9d804afba9346844a3a419e3ca
+size 7569