metadata
language:
- en
- zh
license: apache-2.0
library_name: transformers
tags:
- chat
- conversational
- custom-research
- rp
- roleplay
- tool-use
base_model:
- Qwen/Qwen2.5-14B-Instruct
pipeline_tag: text-generation
model-index:
- name: miscii-14b-1028
results:
- task:
type: text-generation
name: Text Generation
dataset:
name: IFEval (0-Shot)
type: HuggingFaceH4/ifeval
args:
num_few_shot: 0
metrics:
- type: inst_level_strict_acc and prompt_level_strict_acc
value: 82.37
name: strict accuracy
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=sthenno-com/miscii-14b-1028
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: BBH (3-Shot)
type: BBH
args:
num_few_shot: 3
metrics:
- type: acc_norm
value: 49.26
name: normalized accuracy
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=sthenno-com/miscii-14b-1028
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MATH Lvl 5 (4-Shot)
type: hendrycks/competition_math
args:
num_few_shot: 4
metrics:
- type: exact_match
value: 6.34
name: exact match
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=sthenno-com/miscii-14b-1028
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: GPQA (0-shot)
type: Idavidrein/gpqa
args:
num_few_shot: 0
metrics:
- type: acc_norm
value: 14.21
name: acc_norm
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=sthenno-com/miscii-14b-1028
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MuSR (0-shot)
type: TAUR-Lab/MuSR
args:
num_few_shot: 0
metrics:
- type: acc_norm
value: 12
name: acc_norm
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=sthenno-com/miscii-14b-1028
name: Open LLM Leaderboard
- task:
type: text-generation
name: Text Generation
dataset:
name: MMLU-PRO (5-shot)
type: TIGER-Lab/MMLU-Pro
config: main
split: test
args:
num_few_shot: 5
metrics:
- type: acc
value: 46.14
name: accuracy
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=sthenno-com/miscii-14b-1028
name: Open LLM Leaderboard
datasets:
- nvidia/HelpSteer2
- google/Synthetic-Persona-Chat
- custom
miscii-14b-1028
Role-based Instructions
Just use the following text as your system prompt. Note there is NO special-token here, just plain text.
system_prompt: str = (
"""<|context_start|>personas
<|user_persona_start|>statement
{user_persona}<|user_persona_end|>
<|assistant_persona_start|>statement
{assistant_persona}<|assistant_persona_end|><|context_end|>""".format(
user_persona="""I am Miscii.
<optional: personal statement, e.g. I am the designer of Sthenno.>
<optional: additional statements>""",
assistant_persona="""I am Sthenno.
<optional: personal statement, e.g. I speak in Chinese.>
<optional: additional statements>""",
)
)
See Report for miscii-1020 for more details.
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 35.05 |
| IFEval (0-Shot) | 82.37 |
| BBH (3-Shot) | 49.26 |
| MATH Lvl 5 (4-Shot) | 6.34 |
| GPQA (0-shot) | 14.21 |
| MuSR (0-shot) | 12.00 |
| MMLU-PRO (5-shot) | 46.14 |