Upload README.md with huggingface_hub

README.md

@@ -0,0 +1,385 @@
+---
+base_model: OpenLLM-Ro/RoGemma2-9b-Instruct-DPO
+datasets:
+- OpenLLM-Ro/ro_dpo_helpsteer
+language:
+- ro
+license: cc-by-nc-4.0
+tags:
+- llama-cpp
+- gguf-my-repo
+model-index:
+- name: OpenLLM-Ro/RoGemma2-9b-Instruct-DPO-2024-10-09
+  results:
+  - task:
+      type: text-generation
+    dataset:
+      name: RoMT-Bench
+      type: RoMT-Bench
+    metrics:
+    - type: Score
+      value: 6.77
+      name: Score
+    - type: Score
+      value: 7.24
+      name: First turn
+    - type: Score
+      value: 6.3
+      name: Second turn
+  - task:
+      type: text-generation
+    dataset:
+      name: RoCulturaBench
+      type: RoCulturaBench
+    metrics:
+    - type: Score
+      value: 4.83
+      name: Score
+  - task:
+      type: text-generation
+    dataset:
+      name: Romanian_Academic_Benchmarks
+      type: Romanian_Academic_Benchmarks
+    metrics:
+    - type: accuracy
+      value: 59.08
+      name: Average accuracy
+  - task:
+      type: text-generation
+    dataset:
+      name: OpenLLM-Ro/ro_arc_challenge
+      type: OpenLLM-Ro/ro_arc_challenge
+    metrics:
+    - type: accuracy
+      value: 54.1
+      name: Average accuracy
+    - type: accuracy
+      value: 51.59
+      name: 0-shot
+    - type: accuracy
+      value: 50.99
+      name: 1-shot
+    - type: accuracy
+      value: 53.47
+      name: 3-shot
+    - type: accuracy
+      value: 54.84
+      name: 5-shot
+    - type: accuracy
+      value: 58.1
+      name: 10-shot
+    - type: accuracy
+      value: 55.61
+      name: 25-shot
+  - task:
+      type: text-generation
+    dataset:
+      name: OpenLLM-Ro/ro_mmlu
+      type: OpenLLM-Ro/ro_mmlu
+    metrics:
+    - type: accuracy
+      value: 63.41
+      name: Average accuracy
+    - type: accuracy
+      value: 62.15
+      name: 0-shot
+    - type: accuracy
+      value: 62.78
+      name: 1-shot
+    - type: accuracy
+      value: 64.27
+      name: 3-shot
+    - type: accuracy
+      value: 64.43
+      name: 5-shot
+  - task:
+      type: text-generation
+    dataset:
+      name: OpenLLM-Ro/ro_winogrande
+      type: OpenLLM-Ro/ro_winogrande
+    metrics:
+    - type: accuracy
+      value: 70.02
+      name: Average accuracy
+    - type: accuracy
+      value: 66.69
+      name: 0-shot
+    - type: accuracy
+      value: 68.82
+      name: 1-shot
+    - type: accuracy
+      value: 71.82
+      name: 3-shot
+    - type: accuracy
+      value: 72.77
+      name: 5-shot
+  - task:
+      type: text-generation
+    dataset:
+      name: OpenLLM-Ro/ro_hellaswag
+      type: OpenLLM-Ro/ro_hellaswag
+    metrics:
+    - type: accuracy
+      value: 59.35
+      name: Average accuracy
+    - type: accuracy
+      value: 56.98
+      name: 0-shot
+    - type: accuracy
+      value: 57.73
+      name: 1-shot
+    - type: accuracy
+      value: 59.29
+      name: 3-shot
+    - type: accuracy
+      value: 60.7
+      name: 5-shot
+    - type: accuracy
+      value: 62.03
+      name: 10-shot
+  - task:
+      type: text-generation
+    dataset:
+      name: OpenLLM-Ro/ro_gsm8k
+      type: OpenLLM-Ro/ro_gsm8k
+    metrics:
+    - type: accuracy
+      value: 57.24
+      name: Average accuracy
+    - type: accuracy
+      value: 46.78
+      name: 1-shot
+    - type: accuracy
+      value: 59.97
+      name: 3-shot
+    - type: accuracy
+      value: 64.97
+      name: 5-shot
+  - task:
+      type: text-generation
+    dataset:
+      name: OpenLLM-Ro/ro_truthfulqa
+      type: OpenLLM-Ro/ro_truthfulqa
+    metrics:
+    - type: accuracy
+      value: 50.39
+      name: Average accuracy
+  - task:
+      type: text-generation
+    dataset:
+      name: LaRoSeDa_binary
+      type: LaRoSeDa_binary
+    metrics:
+    - type: macro-f1
+      value: 97.74
+      name: Average macro-f1
+    - type: macro-f1
+      value: 97.3
+      name: 0-shot
+    - type: macro-f1
+      value: 97.5
+      name: 1-shot
+    - type: macro-f1
+      value: 97.83
+      name: 3-shot
+    - type: macro-f1
+      value: 98.33
+      name: 5-shot
+  - task:
+      type: text-generation
+    dataset:
+      name: LaRoSeDa_multiclass
+      type: LaRoSeDa_multiclass
+    metrics:
+    - type: macro-f1
+      value: 67.4
+      name: Average macro-f1
+    - type: macro-f1
+      value: 59.3
+      name: 0-shot
+    - type: macro-f1
+      value: 65.52
+      name: 1-shot
+    - type: macro-f1
+      value: 70.94
+      name: 3-shot
+    - type: macro-f1
+      value: 73.85
+      name: 5-shot
+  - task:
+      type: text-generation
+    dataset:
+      name: WMT_EN-RO
+      type: WMT_EN-RO
+    metrics:
+    - type: bleu
+      value: 27.32
+      name: Average bleu
+    - type: bleu
+      value: 17.49
+      name: 0-shot
+    - type: bleu
+      value: 30.33
+      name: 1-shot
+    - type: bleu
+      value: 30.58
+      name: 3-shot
+    - type: bleu
+      value: 30.88
+      name: 5-shot
+  - task:
+      type: text-generation
+    dataset:
+      name: WMT_RO-EN
+      type: WMT_RO-EN
+    metrics:
+    - type: bleu
+      value: 15.96
+      name: Average bleu
+    - type: bleu
+      value: 2.17
+      name: 0-shot
+    - type: bleu
+      value: 10.69
+      name: 1-shot
+    - type: bleu
+      value: 21.68
+      name: 3-shot
+    - type: bleu
+      value: 29.28
+      name: 5-shot
+  - task:
+      type: text-generation
+    dataset:
+      name: XQuAD
+      type: XQuAD
+    metrics:
+    - type: exact_match
+      value: 32.42
+      name: Average exact_match
+    - type: f1
+      value: 58.68
+      name: Average f1
+  - task:
+      type: text-generation
+    dataset:
+      name: STS
+      type: STS
+    metrics:
+    - type: spearman
+      value: 80.82
+      name: Average spearman
+    - type: pearson
+      value: 81.5
+      name: Average pearson
+  - task:
+      type: text-generation
+    dataset:
+      name: XQuAD_EM
+      type: XQuAD_EM
+    metrics:
+    - type: exact_match
+      value: 23.28
+      name: 0-shot
+    - type: exact_match
+      value: 33.45
+      name: 1-shot
+    - type: exact_match
+      value: 34.37
+      name: 3-shot
+    - type: exact_match
+      value: 38.57
+      name: 5-shot
+  - task:
+      type: text-generation
+    dataset:
+      name: XQuAD_F1
+      type: XQuAD_F1
+    metrics:
+    - type: f1
+      value: 47.16
+      name: 0-shot
+    - type: f1
+      value: 60.28
+      name: 1-shot
+    - type: f1
+      value: 62.09
+      name: 3-shot
+    - type: f1
+      value: 65.2
+      name: 5-shot
+  - task:
+      type: text-generation
+    dataset:
+      name: STS_Spearman
+      type: STS_Spearman
+    metrics:
+    - type: spearman
+      value: 75.34
+      name: 1-shot
+    - type: spearman
+      value: 82.71
+      name: 3-shot
+    - type: spearman
+      value: 84.41
+      name: 5-shot
+  - task:
+      type: text-generation
+    dataset:
+      name: STS_Pearson
+      type: STS_Pearson
+    metrics:
+    - type: pearson
+      value: 77.97
+      name: 1-shot
+    - type: pearson
+      value: 82.49
+      name: 3-shot
+    - type: pearson
+      value: 84.05
+      name: 5-shot
+---
+
+# code380/RoGemma2-9b-Instruct-DPO-Q4_K_M-GGUF
+This model was converted to GGUF format from [`OpenLLM-Ro/RoGemma2-9b-Instruct-DPO`](https://huggingface.co/OpenLLM-Ro/RoGemma2-9b-Instruct-DPO) using llama.cpp via the ggml.ai's [GGUF-my-repo](https://huggingface.co/spaces/ggml-org/gguf-my-repo) space.
+Refer to the [original model card](https://huggingface.co/OpenLLM-Ro/RoGemma2-9b-Instruct-DPO) for more details on the model.
+
+## Use with llama.cpp
+Install llama.cpp through brew (works on Mac and Linux)
+
+```bash
+brew install llama.cpp
+
+```
+Invoke the llama.cpp server or the CLI.
+
+### CLI:
+```bash
+llama-cli --hf-repo code380/RoGemma2-9b-Instruct-DPO-Q4_K_M-GGUF --hf-file rogemma2-9b-instruct-dpo-q4_k_m.gguf -p "The meaning to life and the universe is"
+```
+
+### Server:
+```bash
+llama-server --hf-repo code380/RoGemma2-9b-Instruct-DPO-Q4_K_M-GGUF --hf-file rogemma2-9b-instruct-dpo-q4_k_m.gguf -c 2048
+```
+
+Note: You can also use this checkpoint directly through the [usage steps](https://github.com/ggerganov/llama.cpp?tab=readme-ov-file#usage) listed in the Llama.cpp repo as well.
+
+Step 1: Clone llama.cpp from GitHub.
+```
+git clone https://github.com/ggerganov/llama.cpp
+```
+
+Step 2: Move into the llama.cpp folder and build it with `LLAMA_CURL=1` flag along with other hardware-specific flags (for ex: LLAMA_CUDA=1 for Nvidia GPUs on Linux).
+```
+cd llama.cpp && LLAMA_CURL=1 make
+```
+
+Step 3: Run inference through the main binary.
+```
+./llama-cli --hf-repo code380/RoGemma2-9b-Instruct-DPO-Q4_K_M-GGUF --hf-file rogemma2-9b-instruct-dpo-q4_k_m.gguf -p "The meaning to life and the universe is"
+```
+or 
+```
+./llama-server --hf-repo code380/RoGemma2-9b-Instruct-DPO-Q4_K_M-GGUF --hf-file rogemma2-9b-instruct-dpo-q4_k_m.gguf -c 2048
+```