metadata
license: other
library_name: transformers
tags:
- mergekit
- merge
base_model:
- 01-ai/Yi-9B
license_name: yi-license
license_link: LICENSE
model-index:
- name: bigyi-15b
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: 20.94
name: strict accuracy
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=abacusai/bigyi-15b
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: 19.94
name: normalized accuracy
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=abacusai/bigyi-15b
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: 2.34
name: exact match
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=abacusai/bigyi-15b
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: 7.94
name: acc_norm
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=abacusai/bigyi-15b
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: 4.29
name: acc_norm
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=abacusai/bigyi-15b
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: 22.25
name: accuracy
source:
url: >-
https://huggingface.co/spaces/open-llm-leaderboard/open_llm_leaderboard?query=abacusai/bigyi-15b
name: Open LLM Leaderboard
bigyi-15b
I recently made bigstral-12b and then I saw this new awesome model yi-9b and decided to make an embiggened version.
This is a merge of pre-trained language models created using mergekit.
Bigyi-15b is a base / completion model, so there is no chat template.
It has a 4k context.
Example
Here is a recipe for Mai Tai:\n\n1:
3 parts rum, 2: 3 parts pineapple juice, 3: half a cup of lime juice, 4: 6 to 8 fresh or frozen pineapple chunks, 5: crushed ice. Mix all ingredients except ice and pour into glasses with ice. Garnish with a pineapple slice.
Here is an implementation of 2-sum in golang:
func twoSum(nums []int, target int) []int {
if len(nums) <= 1 { return nil }
m := map[int] bool{}
for i := range(nums) {{
n = nums[i]
// find the complement of current number in map
comp = target - n
if comp in m { return [m[comp], i+1 ] }
else { m[n] = true }
}}
return nil
}
Merge Details
Merge Method
This model was merged using the passthrough merge method.
Models Merged
The following models were included in the merge:
Configuration
The following YAML configuration was used to produce this model:
dtype: float16
merge_method: passthrough
slices:
- sources:
- layer_range: [0, 12]
model: 01-ai/Yi-9B
- sources:
- layer_range: [6, 18]
model: 01-ai/Yi-9B
- sources:
- layer_range: [12, 24]
model: 01-ai/Yi-9B
- sources:
- layer_range: [18, 30]
model: 01-ai/Yi-9B
- sources:
- layer_range: [24, 36]
model: 01-ai/Yi-9B
- sources:
- layer_range: [30, 42]
model: 01-ai/Yi-9B
- sources:
- layer_range: [36, 48]
model: 01-ai/Yi-9B
Open LLM Leaderboard Evaluation Results
Detailed results can be found here
| Metric | Value |
|---|---|
| Avg. | 12.95 |
| IFEval (0-Shot) | 20.94 |
| BBH (3-Shot) | 19.94 |
| MATH Lvl 5 (4-Shot) | 2.34 |
| GPQA (0-shot) | 7.94 |
| MuSR (0-shot) | 4.29 |
| MMLU-PRO (5-shot) | 22.25 |