--- license: apache-2.0 datasets: - databricks/databricks-dolly-15k model-index: - name: Instruct_Yi-6B_Dolly15K results: - task: type: text-generation name: Text Generation dataset: name: AI2 Reasoning Challenge (25-Shot) type: ai2_arc config: ARC-Challenge split: test args: num_few_shot: 25 metrics: - type: acc_norm value: 54.86 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=HenryJJ/Instruct_Yi-6B_Dolly15K name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: HellaSwag (10-Shot) type: hellaswag split: validation args: num_few_shot: 10 metrics: - type: acc_norm value: 75.87 name: normalized accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=HenryJJ/Instruct_Yi-6B_Dolly15K name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: MMLU (5-Shot) type: cais/mmlu config: all split: test args: num_few_shot: 5 metrics: - type: acc value: 63.37 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=HenryJJ/Instruct_Yi-6B_Dolly15K name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: TruthfulQA (0-shot) type: truthful_qa config: multiple_choice split: validation args: num_few_shot: 0 metrics: - type: mc2 value: 42.84 source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=HenryJJ/Instruct_Yi-6B_Dolly15K name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: Winogrande (5-shot) type: winogrande config: winogrande_xl split: validation args: num_few_shot: 5 metrics: - type: acc value: 74.9 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=HenryJJ/Instruct_Yi-6B_Dolly15K name: Open LLM Leaderboard - task: type: text-generation name: Text Generation dataset: name: GSM8k (5-shot) type: gsm8k config: main split: test args: num_few_shot: 5 metrics: - type: acc value: 29.26 name: accuracy source: url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=HenryJJ/Instruct_Yi-6B_Dolly15K name: Open LLM Leaderboard --- # Instruct_Yi-6B_Dolly15K Fine-tuned from Yi-6B， used Dolly15k for the dataset. 90% for training, 10% validation. Trained for 2.0 epochs using Lora. Trained with 1024 context window. # Model Details * **Trained by**: trained by HenryJJ. * **Model type:** **Instruct_Yi-6B_Dolly15K** is an auto-regressive language model based on the Llama 2 transformer architecture. * **Language(s)**: English * **License for Instruct_Yi-6B_Dolly15K**: apache-2.0 license # Prompting ## Prompt Template With Context <|startoftext|>[INST]{instruction} {context}[/INST]{response}<|endoftext|> ``` <|startoftext|>[INST] Write a 10-line poem about a given topic The topic is about racecars [/INST] ``` ## Prompt Template Without Context ``` <|startoftext|>[INST] Who was the was the second president of the United States? [/INST] ``` # Training script: Fully opensourced at: https://github.com/hengjiUSTC/learn-llm/blob/main/trl_finetune.py. Run on aws g4dn.12xlarge instance for 4 hours. ``` python3 trl_finetune.py --config configs/yi_6b.yml ``` # Dataset Card for Evaluation run of HenryJJ/Instruct_Yi-6B_Dolly15K Dataset automatically created during the evaluation run of model [HenryJJ/Instruct_Yi-6B_Dolly15K](https://huggingface.co/HenryJJ/Instruct_Yi-6B_Dolly15K) on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard). The dataset is composed of 63 configuration, each one coresponding to one of the evaluated task. The dataset has been created from 1 run(s). Each run can be found as a specific split in each configuration, the split being named using the timestamp of the run.The "train" split is always pointing to the latest results. An additional configuration "results" store all the aggregated results of the run (and is used to compute and display the aggregated metrics on the [Open LLM Leaderboard](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard)). To load the details from a run, you can for instance do the following: ```python from datasets import load_dataset data = load_dataset("open-llm-leaderboard/details_HenryJJ__Instruct_Yi-6B_Dolly15K", "harness_winogrande_5", split="train") ``` ## Latest results These are the [latest results from run 2024-01-06T09:45:44.755529](https://huggingface.co/datasets/open-llm-leaderboard/details_HenryJJ__Instruct_Yi-6B_Dolly15K/blob/main/results_2024-01-06T09-45-44.755529.json)(note that their might be results for other tasks in the repos if successive evals didn't cover the same tasks. You find each in the results and the "latest" split for each eval): ```python { "all": { "acc": 0.6267070831158695, "acc_stderr": 0.03222713761046951, "acc_norm": 0.6343965374667763, "acc_norm_stderr": 0.032887983229700546, "mc1": 0.28886168910648713, "mc1_stderr": 0.01586634640138431, "mc2": 0.42839602626744816, "mc2_stderr": 0.014270024501714959 }, "harness|arc:challenge|25": { "acc": 0.5, "acc_stderr": 0.014611390804670088, "acc_norm": 0.5486348122866894, "acc_norm_stderr": 0.014542104569955265 }, "harness|hellaswag|10": { "acc": 0.5654252141007767, "acc_stderr": 0.004946879874422681, "acc_norm": 0.7587134037044413, "acc_norm_stderr": 0.00426989301158892 }, "harness|hendrycksTest-abstract_algebra|5": { "acc": 0.35, "acc_stderr": 0.0479372485441102, "acc_norm": 0.35, "acc_norm_stderr": 0.0479372485441102 }, "harness|hendrycksTest-anatomy|5": { "acc": 0.562962962962963, "acc_stderr": 0.04284958639753401, "acc_norm": 0.562962962962963, "acc_norm_stderr": 0.04284958639753401 }, "harness|hendrycksTest-astronomy|5": { "acc": 0.6776315789473685, "acc_stderr": 0.03803510248351585, "acc_norm": 0.6776315789473685, "acc_norm_stderr": 0.03803510248351585 }, "harness|hendrycksTest-business_ethics|5": { "acc": 0.7, "acc_stderr": 0.046056618647183814, "acc_norm": 0.7, "acc_norm_stderr": 0.046056618647183814 }, "harness|hendrycksTest-clinical_knowledge|5": { "acc": 0.690566037735849, "acc_stderr": 0.028450154794118637, "acc_norm": 0.690566037735849, "acc_norm_stderr": 0.028450154794118637 }, "harness|hendrycksTest-college_biology|5": { "acc": 0.6666666666666666, "acc_stderr": 0.039420826399272135, "acc_norm": 0.6666666666666666, "acc_norm_stderr": 0.039420826399272135 }, "harness|hendrycksTest-college_chemistry|5": { "acc": 0.41, "acc_stderr": 0.049431107042371025, "acc_norm": 0.41, "acc_norm_stderr": 0.049431107042371025 }, "harness|hendrycksTest-college_computer_science|5": { "acc": 0.44, "acc_stderr": 0.04988876515698589, "acc_norm": 0.44, "acc_norm_stderr": 0.04988876515698589 }, "harness|hendrycksTest-college_mathematics|5": { "acc": 0.36, "acc_stderr": 0.04824181513244218, "acc_norm": 0.36, "acc_norm_stderr": 0.04824181513244218 }, "harness|hendrycksTest-college_medicine|5": { "acc": 0.6069364161849711, "acc_stderr": 0.03724249595817731, "acc_norm": 0.6069364161849711, "acc_norm_stderr": 0.03724249595817731 }, "harness|hendrycksTest-college_physics|5": { "acc": 0.3235294117647059, "acc_stderr": 0.04655010411319617, "acc_norm": 0.3235294117647059, "acc_norm_stderr": 0.04655010411319617 }, "harness|hendrycksTest-computer_security|5": { "acc": 0.77, "acc_stderr": 0.04229525846816507, "acc_norm": 0.77, "acc_norm_stderr": 0.04229525846816507 }, "harness|hendrycksTest-conceptual_physics|5": { "acc": 0.6212765957446809, "acc_stderr": 0.03170995606040655, "acc_norm": 0.6212765957446809, "acc_norm_stderr": 0.03170995606040655 }, "harness|hendrycksTest-econometrics|5": { "acc": 0.35964912280701755, "acc_stderr": 0.045144961328736334, "acc_norm": 0.35964912280701755, "acc_norm_stderr": 0.045144961328736334 }, "harness|hendrycksTest-electrical_engineering|5": { "acc": 0.6482758620689655, "acc_stderr": 0.0397923663749741, "acc_norm": 0.6482758620689655, "acc_norm_stderr": 0.0397923663749741 }, "harness|hendrycksTest-elementary_mathematics|5": { "acc": 0.4470899470899471, "acc_stderr": 0.02560672399577703, "acc_norm": 0.4470899470899471, "acc_norm_stderr": 0.02560672399577703 }, "harness|hendrycksTest-formal_logic|5": { "acc": 0.38095238095238093, "acc_stderr": 0.04343525428949098, "acc_norm": 0.38095238095238093, "acc_norm_stderr": 0.04343525428949098 }, "harness|hendrycksTest-global_facts|5": { "acc": 0.4, "acc_stderr": 0.04923659639173309, "acc_norm": 0.4, "acc_norm_stderr": 0.04923659639173309 }, "harness|hendrycksTest-high_school_biology|5": { "acc": 0.7774193548387097, "acc_stderr": 0.023664216671642525, "acc_norm": 0.7774193548387097, "acc_norm_stderr": 0.023664216671642525 }, "harness|hendrycksTest-high_school_chemistry|5": { "acc": 0.4975369458128079, "acc_stderr": 0.03517945038691063, "acc_norm": 0.4975369458128079, "acc_norm_stderr": 0.03517945038691063 }, "harness|hendrycksTest-high_school_computer_science|5": { "acc": 0.64, "acc_stderr": 0.04824181513244218, "acc_norm": 0.64, "acc_norm_stderr": 0.04824181513244218 }, "harness|hendrycksTest-high_school_european_history|5": { "acc": 0.7393939393939394, "acc_stderr": 0.034277431758165236, "acc_norm": 0.7393939393939394, "acc_norm_stderr": 0.034277431758165236 }, "harness|hendrycksTest-high_school_geography|5": { "acc": 0.8181818181818182, "acc_stderr": 0.0274796030105388, "acc_norm": 0.8181818181818182, "acc_norm_stderr": 0.0274796030105388 }, "harness|hendrycksTest-high_school_government_and_politics|5": { "acc": 0.9015544041450777, "acc_stderr": 0.021500249576033456, "acc_norm": 0.9015544041450777, "acc_norm_stderr": 0.021500249576033456 }, "harness|hendrycksTest-high_school_macroeconomics|5": { "acc": 0.617948717948718, "acc_stderr": 0.02463554916390823, "acc_norm": 0.617948717948718, "acc_norm_stderr": 0.02463554916390823 }, "harness|hendrycksTest-high_school_mathematics|5": { "acc": 0.31851851851851853, "acc_stderr": 0.028406533090608463, "acc_norm": 0.31851851851851853, "acc_norm_stderr": 0.028406533090608463 }, "harness|hendrycksTest-high_school_microeconomics|5": { "acc": 0.7647058823529411, "acc_stderr": 0.027553614467863797, "acc_norm": 0.7647058823529411, "acc_norm_stderr": 0.027553614467863797 }, "harness|hendrycksTest-high_school_physics|5": { "acc": 0.36423841059602646, "acc_stderr": 0.03929111781242742, "acc_norm": 0.36423841059602646, "acc_norm_stderr": 0.03929111781242742 }, "harness|hendrycksTest-high_school_psychology|5": { "acc": 0.8348623853211009, "acc_stderr": 0.01591955782997604, "acc_norm": 0.8348623853211009, "acc_norm_stderr": 0.01591955782997604 }, "harness|hendrycksTest-high_school_statistics|5": { "acc": 0.5694444444444444, "acc_stderr": 0.03376922151252335, "acc_norm": 0.5694444444444444, "acc_norm_stderr": 0.03376922151252335 }, "harness|hendrycksTest-high_school_us_history|5": { "acc": 0.8088235294117647, "acc_stderr": 0.027599174300640766, "acc_norm": 0.8088235294117647, "acc_norm_stderr": 0.027599174300640766 }, "harness|hendrycksTest-high_school_world_history|5": { "acc": 0.7932489451476793, "acc_stderr": 0.026361651668389094, "acc_norm": 0.7932489451476793, "acc_norm_stderr": 0.026361651668389094 }, "harness|hendrycksTest-human_aging|5": { "acc": 0.695067264573991, "acc_stderr": 0.030898610882477515, "acc_norm": 0.695067264573991, "acc_norm_stderr": 0.030898610882477515 }, "harness|hendrycksTest-human_sexuality|5": { "acc": 0.7480916030534351, "acc_stderr": 0.03807387116306085, "acc_norm": 0.7480916030534351, "acc_norm_stderr": 0.03807387116306085 }, "harness|hendrycksTest-international_law|5": { "acc": 0.7768595041322314, "acc_stderr": 0.03800754475228733, "acc_norm": 0.7768595041322314, "acc_norm_stderr": 0.03800754475228733 }, "harness|hendrycksTest-jurisprudence|5": { "acc": 0.7777777777777778, "acc_stderr": 0.040191074725573483, "acc_norm": 0.7777777777777778, "acc_norm_stderr": 0.040191074725573483 }, "harness|hendrycksTest-logical_fallacies|5": { "acc": 0.7852760736196319, "acc_stderr": 0.03226219377286775, "acc_norm": 0.7852760736196319, "acc_norm_stderr": 0.03226219377286775 }, "harness|hendrycksTest-machine_learning|5": { "acc": 0.4375, "acc_stderr": 0.04708567521880525, "acc_norm": 0.4375, "acc_norm_stderr": 0.04708567521880525 }, "harness|hendrycksTest-management|5": { "acc": 0.8155339805825242, "acc_stderr": 0.03840423627288276, "acc_norm": 0.8155339805825242, "acc_norm_stderr": 0.03840423627288276 }, "harness|hendrycksTest-marketing|5": { "acc": 0.8974358974358975, "acc_stderr": 0.01987565502786744, "acc_norm": 0.8974358974358975, "acc_norm_stderr": 0.01987565502786744 }, "harness|hendrycksTest-medical_genetics|5": { "acc": 0.76, "acc_stderr": 0.042923469599092816, "acc_norm": 0.76, "acc_norm_stderr": 0.042923469599092816 }, "harness|hendrycksTest-miscellaneous|5": { "acc": 0.8007662835249042, "acc_stderr": 0.014283378044296417, "acc_norm": 0.8007662835249042, "acc_norm_stderr": 0.014283378044296417 }, "harness|hendrycksTest-moral_disputes|5": { "acc": 0.708092485549133, "acc_stderr": 0.024476994076247333, "acc_norm": 0.708092485549133, "acc_norm_stderr": 0.024476994076247333 }, "harness|hendrycksTest-moral_scenarios|5": { "acc": 0.33519553072625696, "acc_stderr": 0.015788007190185884, "acc_norm": 0.33519553072625696, "acc_norm_stderr": 0.015788007190185884 }, "harness|hendrycksTest-nutrition|5": { "acc": 0.7222222222222222, "acc_stderr": 0.025646863097137897, "acc_norm": 0.7222222222222222, "acc_norm_stderr": 0.025646863097137897 }, "harness|hendrycksTest-philosophy|5": { "acc": 0.6913183279742765, "acc_stderr": 0.026236965881153262, "acc_norm": 0.6913183279742765, "acc_norm_stderr": 0.026236965881153262 }, "harness|hendrycksTest-prehistory|5": { "acc": 0.7191358024691358, "acc_stderr": 0.025006469755799208, "acc_norm": 0.7191358024691358, "acc_norm_stderr": 0.025006469755799208 }, "harness|hendrycksTest-professional_accounting|5": { "acc": 0.48226950354609927, "acc_stderr": 0.02980873964223777, "acc_norm": 0.48226950354609927, "acc_norm_stderr": 0.02980873964223777 }, "harness|hendrycksTest-professional_law|5": { "acc": 0.4876140808344198, "acc_stderr": 0.012766317315473565, "acc_norm": 0.4876140808344198, "acc_norm_stderr": 0.012766317315473565 }, "harness|hendrycksTest-professional_medicine|5": { "acc": 0.6213235294117647, "acc_stderr": 0.02946513363977613, "acc_norm": 0.6213235294117647, "acc_norm_stderr": 0.02946513363977613 }, "harness|hendrycksTest-professional_psychology|5": { "acc": 0.6568627450980392, "acc_stderr": 0.019206606848825365, "acc_norm": 0.6568627450980392, "acc_norm_stderr": 0.019206606848825365 }, "harness|hendrycksTest-public_relations|5": { "acc": 0.6909090909090909, "acc_stderr": 0.044262946482000985, "acc_norm": 0.6909090909090909, "acc_norm_stderr": 0.044262946482000985 }, "harness|hendrycksTest-security_studies|5": { "acc": 0.7306122448979592, "acc_stderr": 0.02840125202902294, "acc_norm": 0.7306122448979592, "acc_norm_stderr": 0.02840125202902294 }, "harness|hendrycksTest-sociology|5": { "acc": 0.8159203980099502, "acc_stderr": 0.027403859410786862, "acc_norm": 0.8159203980099502, "acc_norm_stderr": 0.027403859410786862 }, "harness|hendrycksTest-us_foreign_policy|5": { "acc": 0.84, "acc_stderr": 0.03684529491774708, "acc_norm": 0.84, "acc_norm_stderr": 0.03684529491774708 }, "harness|hendrycksTest-virology|5": { "acc": 0.4578313253012048, "acc_stderr": 0.0387862677100236, "acc_norm": 0.4578313253012048, "acc_norm_stderr": 0.0387862677100236 }, "harness|hendrycksTest-world_religions|5": { "acc": 0.8070175438596491, "acc_stderr": 0.030267457554898458, "acc_norm": 0.8070175438596491, "acc_norm_stderr": 0.030267457554898458 }, "harness|truthfulqa:mc|0": { "mc1": 0.28886168910648713, "mc1_stderr": 0.01586634640138431, "mc2": 0.42839602626744816, "mc2_stderr": 0.014270024501714959 }, "harness|winogrande|5": { "acc": 0.7490134175217048, "acc_stderr": 0.012185776220516148 }, "harness|gsm8k|5": { "acc": 0.2926459438968916, "acc_stderr": 0.012532334368242888 } } ``` # [Open LLM Leaderboard Evaluation Results](https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard) Detailed results can be found [here](https://huggingface.co/datasets/open-llm-leaderboard/details_HenryJJ__Instruct_Yi-6B_Dolly15K) | Metric |Value| |---------------------------------|----:| |Avg. |56.85| |AI2 Reasoning Challenge (25-Shot)|54.86| |HellaSwag (10-Shot) |75.87| |MMLU (5-Shot) |63.37| |TruthfulQA (0-shot) |42.84| |Winogrande (5-shot) |74.90| |GSM8k (5-shot) |29.26|