You need to agree to share your contact information to access this model

This repository is publicly accessible, but you have to accept the conditions to access its files and content.

Log in or Sign Up to review the conditions and access this model content.

Model Card for Model lianghsun/Llama-3.2-Taiwan-Legal-3B-Instruct

image/png

ๅŸบๆ–ผ lianghsun/Llama-3.2-Taiwan-3B ๆจกๅž‹๏ผŒ้€้Žไธญ่ฏๆฐ‘ๅœ‹ๅฐ็ฃๆณ•ๅพ‹ๆขๆ–‡ๅŠๅˆคๆฑบๆ›ธ็ญ‰็›ธ้—œ่ณ‡ๆ–™้›†้€ฒ่กŒๅพฎ่ชฟใ€‚

Model Update History

Update Date Model Version Key Changes
2024-11-13 v2024.11.13 Fine-tuned version v2024.11.13 of lianghsun/Llama-3.2-Taiwan-3B. This instruction version initiates experimental integration of non-Chinese instructions to enhance the modelโ€™s robustness and mitigate risks of overfitting.
2024-11-06 v2024.11.6 Starting with this release, fine-tuning is based on the foundation model lianghsun/Llama-3.2-Taiwan-3B v2024.10.27 , and versioning has been updated to use the YYYY-mm-dd format.
2024-10-17 v1.1.0 (v2024.10.17) (Model collapsed ๐Ÿ’ฅ) Experimental fine-tuning on v1.0.0 with added legal code data from the Republic of China (Taiwan)
2024-10-10 v1.0.0 (v2024.10.10) Full model training completed, but missing legal code data for the Republic of China (Taiwan)
2024-09-27 v0.1.0 (v2024.09.27) Model v0.1.0 released, but training was interrupted after 3 epochs due to lack of compute resources

Model Details

Model Description

ๅŸบๆ–ผ lianghsun/Llama-3.2-Taiwan-3B ๆจกๅž‹๏ผŒๆญคๅพฎ่ชฟ้Ž็จ‹ไฝฟ็”จไบ†ไพ†่‡ชไธญ่ฏๆฐ‘ๅœ‹ๅฐ็ฃ็š„ๆณ•ๅพ‹ๆขๆ–‡่ˆ‡็›ธ้—œๅˆคๆฑบๆ›ธ่ณ‡ๆ–™้›†๏ผŒไปฅๆๅ‡ๆจกๅž‹ๅœจๆณ•ๅพ‹้ ˜ๅŸŸ็š„ๅฐˆๆฅญ็Ÿฅ่ญ˜่ˆ‡ๆ‡‰็”จ่ƒฝๅŠ›ใ€‚้€™ไบ›่ณ‡ๆ–™้›†ๆถต่“‹ไบ†ๆณ•ๅพ‹ๆขๆ–‡็š„็ตๆง‹ใ€ๅˆคๆฑบๆ›ธ็š„ๆ ผๅผ๏ผŒๆณ•ๅบญไธŠๅธธ่ฆ‹็š„ๆณ•ๅพ‹่ชž่จ€่ˆ‡่ก“่ชž๏ผŒไธฆๅŒ…ๅซไบ†้ƒจๅˆ†ๆณ•ๅพ‹่ณ‡ๆ–™็ง‘ๅญธไปปๅ‹™็š„ๆ‡‰็”จ๏ผŒไฝฟๆจกๅž‹่ƒฝๅค ๆ›ดๆบ–็ขบๅœฐ็†่งฃๅ’Œ่™•็†่ˆ‡ๅฐ็ฃๆณ•ๅพ‹้ซ”็ณป็›ธ้—œ็š„ๅ•้กŒใ€‚็ถ“้Ž้€™ไบ›ๅพฎ่ชฟ๏ผŒๆจกๅž‹ๅฐ‡่ƒฝๅค ๆ›ดๅฅฝๅœฐ็‚บๆณ•ๅพ‹ๅฐˆๆฅญไบบๅฃซๆไพ›ๅนซๅŠฉ๏ผŒไธฆๅœจๅฐ็ฃๆณ•ๅˆถๆก†ๆžถๅ…งๆไพ›ๆ›ด็ฒพๆบ–็š„ๅ›žๆ‡‰่ˆ‡ๅปบ่ญฐใ€‚

  • Developed by: Huang Liang Hsun
  • Model type: LlamaForCausalLM
  • Language(s) (NLP): ไธป่ฆ่™•็†็น้ซ”ไธญๆ–‡๏ผˆzh-tw๏ผ‰๏ผŒ้‡ๅฐไธญ่ฏๆฐ‘ๅœ‹ๅฐ็ฃ็š„ๆณ•ๅพ‹็”จ่ชž่ˆ‡ๅˆคๆฑบๆ›ธ้€ฒ่กŒๅพฎ่ชฟใ€‚
  • License: llama3.2
  • Finetuned from model: lianghsun/Llama-3.2-Taiwan-3B

Model Sources

Uses

Direct Use

ๆญคๆจกๅž‹ๅฏไปฅ็›ดๆŽฅ็”จๆ–ผ็†่งฃๅ’Œ็”Ÿๆˆ็น้ซ”ไธญๆ–‡ๆณ•ๅพ‹ๆ–‡ๆœฌ๏ผŒ้ฉๅˆ้œ€่ฆ่™•็†ๅฐ็ฃๆณ•ๅพ‹็›ธ้—œๅ•้กŒ็š„ๆ‡‰็”จๅ ดๆ™ฏใ€‚ๆจกๅž‹้ ่จญ็š„ๆŒ‡ไปคๅ’Œๅ›žๆ‡‰่ƒฝๅค ๆœ‰ๆ•ˆๆไพ›ๆณ•ๅพ‹่ณ‡่จŠใ€้‡ๆธ…ๆณ•ๅพ‹ๆขๆ–‡ใ€ไธฆ็”Ÿๆˆ็ฌฆๅˆๆณ•ๅพ‹ๅฐˆๆฅญ็š„ๅ›žๆ‡‰ใ€‚ๅ…ถ็›ดๆŽฅไฝฟ็”จ็ฏ„ๅœๅŒ…ๆ‹ฌไฝ†ไธ้™ๆ–ผๆณ•ๅพ‹่ณ‡่จŠๆŸฅ่ฉขใ€ๆณ•ๅพ‹ๆ–‡ๆœฌๆ‘˜่ฆใ€ๅ’ŒๅŸบๆœฌ็š„ๆณ•ๆขๅฐ่ฉฑใ€‚

Downstream Use

็ถ“้Žๅพฎ่ชฟๅพŒ๏ผŒ่ฉฒๆจกๅž‹ๅฏ็”จๆ–ผๆ›ดๅ…ท้ซ”็š„ๆณ•ๅพ‹ไปปๅ‹™๏ผŒๅฆ‚่‡ชๅ‹•ๅˆคๆฑบๆ›ธๅˆ†ๆžใ€ๆณ•ๅพ‹ๅฏฆ้ซ”่ญ˜ๅˆฅ๏ผˆNER๏ผ‰ใ€ๆณ•่ฆ็ทจ่™Ÿ่ฝ‰ๆ›๏ผŒไปฅๅŠๆณ•ๅพ‹ๅˆ่ฆๅฏฉๆŸฅ่ผ”ๅŠฉใ€‚ๆญคๆจกๅž‹ๅฏไปฅ็„ก็ธซ้›†ๆˆ่‡ณๆณ•ๅพ‹ๆ•ธๆ“š็ง‘ๅญธๆ‡‰็”จๆˆ–ๆณ•ๅพ‹ๆŠ€่ก“๏ผˆLegalTech๏ผ‰็ณป็ตฑไธญ๏ผŒๅนซๅŠฉๆณ•ๅพ‹ๅฐˆๆฅญไบบๅฃซๆˆ–ไผๆฅญๆๅ‡ๅทฅไฝœๆ•ˆ็އใ€‚

Out-of-Scope Use

่ฉฒๆจกๅž‹ไธฆไธ้ฉ็”จๆ–ผ้žๆณ•ๅพ‹็›ธ้—œ้ ˜ๅŸŸ็š„็”Ÿๆˆไปปๅ‹™๏ผŒไธ”ไธๆ‡‰็”จๆ–ผ้€ฒ่กŒๅฏ่ƒฝๆถ‰ๅŠ่ชคๅฐŽๆˆ–้Œฏ่ชค็š„ๆณ•ๅพ‹ๅปบ่ญฐ๏ผŒๅฐคๅ…ถๆ˜ฏๅœจๆœช็ถ“ๅฐˆๆฅญๅฏฉๆŸฅ็š„ๆƒ…ๆณไธ‹ใ€‚้ฟๅ…ๅฐ‡ๆจกๅž‹็”จๆ–ผๆœช็ถ“ๆŽˆๆฌŠๆˆ–้žๆณ•็”จ้€”๏ผŒๅฆ‚็”Ÿๆˆๅ…ท็ˆญ่ญฐๆ€งๆˆ–ๅ…ทๅ่ฆ‹็š„ๆณ•ๅพ‹ๅปบ่ญฐใ€‚

Bias, Risks, and Limitations

ๆจกๅž‹ๅœจ็”Ÿๆˆๆณ•ๅพ‹ๆขๆ–‡ๅ’Œๅˆคๆฑบๆ›ธๅ…งๅฎนๆ™‚๏ผŒๅฏ่ƒฝๆœƒ็”Ÿๆˆ่™›ๆง‹ๆˆ–ไธๅญ˜ๅœจ็š„ๆณ•ๆขๆˆ–ๅˆคๆฑบๆ›ธๅ…งๅฎน๏ผŒ้€™ๆ˜ฏๆจกๅž‹็š„ๅ…งๅœจ้™ๅˆถไน‹ไธ€ใ€‚ไฝฟ็”จ่€…ๅœจๅƒ่€ƒ้€™ไบ›่ณ‡ๆ–™ๆ™‚๏ผŒๆ‡‰่ฌนๆ…ŽๆชขๆŸฅ็”Ÿๆˆ็š„ๅ…งๅฎน๏ผŒไธฆ้ฟๅ…ๅฐ‡ๆจกๅž‹่ผธๅ‡บ่ฆ–็‚บๆณ•ๅพ‹ไพๆ“šใ€‚ๅปบ่ญฐๅœจๅฏฆ้š›ๆ‡‰็”จไธญ๏ผŒๅฐ‡ๆจกๅž‹็”Ÿๆˆ็š„็ตๆžœ่ˆ‡ๅฏ้ ็š„ๆณ•ๅพ‹่ฆ‹่งฃๅ’Œไพ†ๆบ้€ฒ่กŒๆฏ”ๅฐ๏ผŒ็ขบไฟๆบ–็ขบๆ€งใ€ๅˆๆณ•ๆ€งๅ’Œ้ฉ็”จๆ€งใ€‚

Recommendations

ๆญคๆจกๅž‹้›–็„ถ็ถ“้Žๆณ•ๅพ‹ๆ–‡ๆœฌ็š„ๅพฎ่ชฟ๏ผŒไฝ†ๅœจๆ–ผๆณ•ๅพ‹ๆ–‡ๆœฌ็š„ๆ•ธ้‡ๅŠๅŸบ็คŽๆจกๅž‹็‚บ SLM๏ผŒๆจกๅž‹่ƒฝๅŠ›ไปๆœ‰ๆฅต้™๏ผŒไฝฟ็”จ่€…ๆ‡‰ๆณจๆ„ไปฅไธ‹้ขจ้šช่ˆ‡้™ๅˆถ๏ผš

  1. ๅ่ฆ‹้ขจ้šช๏ผš ๆจกๅž‹ๅฏ่ƒฝๆœƒๅๆ˜ ๅ…ถ่จ“็ทด่ณ‡ๆ–™ไธญ็š„ๆฝ›ๅœจๅ่ฆ‹ใ€‚็”ฑๆ–ผๆณ•ๅพ‹ๆ–‡ๆœฌ็š„็‰นๅฎšๆ€ง๏ผŒๆจกๅž‹ๅฏ่ƒฝๆ›ด็†Ÿๆ‚‰ๆŸไบ›ๆณ•่ฆใ€ๆขๆ–‡ๆˆ–ๅˆคๆฑบๆกˆไพ‹๏ผŒ่€Œๅœจๅ…ถไป–้ ˜ๅŸŸ่กจ็พ่ผƒๅผฑใ€‚็‰นๅˆฅๆ˜ฏๅœจ่™•็†ไธๅธธ่ฆ‹็š„ๆณ•ๅพ‹ๅ•้กŒๆˆ–ๆœช่ขซ่จ“็ทด้Ž็š„ๆ–ฐๆณ•่ฆๆ™‚๏ผŒๆจกๅž‹็š„่ผธๅ‡บๅฏ่ƒฝๆœƒๅธถๆœ‰ๅ่ฆ‹ใ€‚

  2. ๆŠ€่ก“้™ๅˆถ๏ผš ้›–็„ถๆจกๅž‹่ƒฝๅค ่™•็†ๅคง้ƒจๅˆ†็š„ๆณ•ๅพ‹ๆ–‡ๆœฌ๏ผŒไฝ†ๅฐๆ–ผ็ตๆง‹ๆฅตๅ…ถ่ค‡้›œๆˆ–่ชž่จ€ๆจกๆฃฑๅ…ฉๅฏ็š„ๆณ•ๅพ‹ๆขๆ–‡๏ผŒๆจกๅž‹ๅฏ่ƒฝ็„กๆณ•็”ข็”Ÿ็ฒพ็ขบ็š„ๅ›ž็ญ”ใ€‚ไฝฟ็”จ่€…ๆ‡‰้ฟๅ…ๅฎŒๅ…จไพ่ณดๆจกๅž‹็š„่ผธๅ‡บ๏ผŒๅฐคๅ…ถๅœจๆณ•ๅพ‹ๆฑบ็ญ–้Ž็จ‹ไธญ๏ผŒๅปบ่ญฐ้€ฒ่กŒ้กๅค–็š„ๅฐˆๆฅญๆชขๆŸฅใ€‚

  3. ๆณ•ๅพ‹่ฒฌไปป๏ผš ๆจกๅž‹ไธฆ้žๅฐˆๆฅญๆณ•ๅพ‹้กงๅ•๏ผŒๅ› ๆญคๅ…ถ็”Ÿๆˆ็š„ๅ›žๆ‡‰ไธๆ‡‰่ขซ่ฆ–็‚บๆญฃ็ขบ็š„ๆณ•ๅพ‹ๅปบ่ญฐใ€‚ไฝฟ็”จ่€…ๆ‡‰็ขบไฟๅœจ็†ๆ€งไธ”ๅฐˆๆฅญ่ƒŒๆ™ฏไธ‹้€ฒ่กŒๆจกๅž‹็š„ๆ‡‰็”จ๏ผŒไธฆ้ฟๅ…ๅœจ้—œ้ตๆฑบ็ญ–ไธญ้Žๅบฆไพ่ณดๆจกๅž‹ใ€‚

  4. ่ชค็”จ้ขจ้šช๏ผš ไธ็•ถไฝฟ็”จๆจกๅž‹้€ฒ่กŒ้Œฏ่ชคๆˆ–่ชคๅฐŽๆ€ง็š„ๆณ•ๅพ‹ๅปบ่ญฐ๏ผŒๅฏ่ƒฝๅฐๅ€‹ไบบๆˆ–ไผๆฅญ้€ ๆˆ่ฒ ้ขๅฝฑ้Ÿฟใ€‚ไฝฟ็”จ่€…ๆ‡‰่ฌนๆ…Žๆ‡‰็”จๆจกๅž‹ๆ–ผๅˆ่ฆๆˆ–ๆณ•ๅพ‹็›ธ้—œไปปๅ‹™ไธญ๏ผŒไธฆไฟๆŒๅฐๅ…ถ่ผธๅ‡บ็š„ๆชข่ฆ–ๅŠๆ กๆญฃใ€‚

็‚บไบ†ๆธ›ๅฐ‘้€™ไบ›้ขจ้šช๏ผŒๅปบ่ญฐไฝฟ็”จ่€…ๅœจๆ‡‰็”จๆจกๅž‹่ผธๅ‡บๆ™‚้€ฒ่กŒไบŒๆฌกๆชขๆŸฅ๏ผŒ็‰นๅˆฅๆ˜ฏๅœจๆถ‰ๅŠๆณ•ๅพ‹ๆฑบ็ญ–็š„ๆƒ…ๅขƒไธญใ€‚ๆœฌๆจกๅž‹็พ้šŽๆฎต็‚บๆไพ›ๆณ•ๅพ‹็ง‘ๆŠ€้ ˜ๅŸŸ้€ฒ่กŒๅคง่ชž่จ€ๆจกๅž‹็ ”็ฉถ๏ผŒไธฆ้žๅ–ไปฃๅฐˆๆฅญๆณ•ๅพ‹ๅทฅไฝœ่€…ไน‹ๅฐˆๆฅญๅปบ่ญฐใ€‚

How to Get Started with the Model

Using vLLM

่ฆไฝฟ็”จ vLLM Docker image ไพ†ๅ•Ÿๅ‹•ๆญคๆจกๅž‹๏ผŒๆ‚จๅฏไปฅๆŒ‰็…งไปฅไธ‹ๆ“ไฝœ๏ผš

docker run --runtime nvidia --gpus all \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HUGGING_FACE_HUB_TOKEN=<secret>" \
    -p 8000:8000 \
    --ipc=host \
    vllm/vllm-openai:latest \
    --model lianghsun/Llama-3.2-Taiwan-Legal-3B-Instruct

่ซ‹ๆณจๆ„๏ผŒๅฆ‚ๆžœๆƒณ่ฆไฝฟ็”จไธๅŒ็‰ˆๆœฌ็š„ checkpoint๏ผŒ่ซ‹ๅŠ ไธŠ --revision <tag_name> ๏ผš

docker run --runtime nvidia --gpus all \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HUGGING_FACE_HUB_TOKEN=<secret>" \
    -p 8000:8000 \
    --ipc=host \
    vllm/vllm-openai:latest \
    --model lianghsun/Llama-3.2-Taiwan-Legal-3B-Instruct --revision <tag_name>

Training Details

Training Data (for v2024.11.13)

Training procedure

Preprocessing

ๆญคๆจกๅž‹ๅœจ v2024.11.06 ็‰ˆๆœฌๅพŒๆ”นๆŽก็”จ lianghsun/Llama-3.2-Taiwan-3B ๅš็‚บๅŸบ็คŽๆจกๅž‹๏ผˆfoundation model๏ผ‰ใ€‚ Tokenizer ไป่ˆ‡ๆœ€ๅŽŸๅ…ˆ็š„ meta-llama/Llama-3.2-3B ็›ธๅŒ๏ผŒๅœจๆœชไพ†็‰ˆๆœฌๅ†่ฆ–ๆณๆ“ดๅ……ไธญๆ–‡ๅญ—ใ€‚

Training hyperparameters (for v2024.11.13)

The following hyperparameters were used during training:

  • learning_rate: (initial lr) 5e-5
  • train_batch_size: 20
  • eval_batch_size: Not specified
  • seed: 42
  • distributed_type: single-node
  • num_devices: 8
  • gradient_accumulation_steps: 16
  • total_train_batch_size: 1,280 (train_batch_size * gradient_accumulation_steps * num_devices)
  • optimizer: adamw_torch_fused
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_steps: 100
  • num_epochs: 3
  • grad_norm: 1.1764454343711086
  • global_step: 65

Speeds, Sizes, Times (for v2024.11.13)

  • Duration: 30m 19s
  • Train runtime: 30m 19s
  • Train samples per second: 1.1764454343711086
  • Train steps per second: 0.036
  • Total training FLOPs: 89,423,735,685,120
  • Train loss: 0.7657

Evaluation

Testing Data, Factors & Metrics

Testing Data

Note: ..(WIP)..

Factors

Note: ..(WIP)..

Metrics

Note: ..(WIP)..

Results

Note: ..(WIP)..

Summary

Note: ..(WIP)..

Model Examination

ๆณ•ๆขๅ›ž่ฆ†

Note: ..(WIP)..

ๅˆคๆฑบๆ›ธๅ…งๅฎน

Note: ..(WIP)..

ๆณ•ๅพ‹ NLP ไปปๅ‹™

Note: ..(WIP)..

Environmental Impact (for v2024.11.13)

  • Hardware Type: 8 x NVIDIA A100 40GB
  • Hours used: 30m 19s
  • Cloud Provider: N/A
  • Compute Region: N/A
  • Carbon Emitted: N/A

Technical Specifications

Model Architecture and Objective

ๆœฌๆจกๅž‹ๅŸบๆ–ผ lianghsun/Llama-3.2-Taiwan-3B ๏ผŒไฝฟ็”จ่‡ชๅ›žๆญธ Transformer ๆžถๆง‹้€ฒ่กŒ่ชž่จ€ๅปบๆจกใ€‚่ฉฒๆจกๅž‹็š„ไธป่ฆ็›ฎๆจ™ๆ˜ฏๆๅ‡ๅฐๅฐ็ฃๆณ•ๅพ‹ๆ–‡ๆœฌ็š„็†่งฃ่ˆ‡็”Ÿๆˆ่ƒฝๅŠ›๏ผŒๅฐคๅ…ถๆ˜ฏ้‡ๅฐๅˆคๆฑบๆ›ธใ€ๆณ•ๆข็š„ๅฐˆๆฅญ่™•็†่ˆ‡ๆ‡‰็”จใ€‚้€้Žๅฐˆ้–€่จญ่จˆ็š„ๆณ•ๅพ‹ๆ–‡ๆœฌ้›†้€ฒ่กŒๅพฎ่ชฟ๏ผŒๆจกๅž‹่ƒฝๆ›ด็ฒพ็ขบๅœฐๅ›ž็ญ”ๆณ•ๅพ‹ๅ•้กŒไธฆๆไพ›็›ธ้—œๅปบ่ญฐใ€‚

Compute Infrastructure

Hardware (for v2024.11.6)

  • 8 x NVIDIA A100 40GB

Software

  • ๅพฎ่ชฟ้Ž็จ‹ไฝฟ็”จไบ† hiyouga/LLaMA-Factory ๆก†ๆžถ้€ฒ่กŒ่จ“็ทดใ€‚

Citation

็„กใ€‚

Glossary

็„กใ€‚

More Information

็ฎ—ๅŠ›

ๅ„˜็ฎกๆˆ‘ๅ€‘ๅทฒๆบ–ๅ‚™ไบ†่จฑๅคš้—œๆ–ผไธญ่ฏๆฐ‘ๅœ‹ๅฐ็ฃๆณ•ๅพ‹้ ˜ๅŸŸ็š„่ณ‡ๆ–™้›†๏ผŒไฝ†็”ฑๆ–ผ็ฎ—ๅŠ›่ณ‡ๆบๆœ‰้™๏ผŒ็„กๆณ•ๅฐ‡ๆ‰€ๆœ‰่ณ‡ๆ–™้›†้€ฒ่กŒๅฎŒๆ•ด่จ“็ทด๏ผˆๆ˜ฏ็š„๏ผŒๆˆ‘ๅ€‘ไธฆๆฒ’ๆœ‰ๅฐ‡ๅ…จ้ƒจ่ณ‡ๆ–™้›†้ƒฝ้€ฒ่กŒ่จ“็ทด๏ผŒๅƒ…ๅ–ๅ‡บ่ขซ่ช็‚บๆœ€ๅŸบ็คŽ็š„ๆณ•ๅพ‹ๆ–‡ๆœฌ๏ผ‰๏ผŒๅฐŽ่‡ดๆจกๅž‹ๅฐšๆœช้”ๅˆฐๆœ€ไฝณ่กจ็พใ€‚ๅ› ๆญค๏ผŒ็›ฎๅ‰็š„ checkpoint ๆ˜ฏๅŸบๆ–ผๆœ‰้™่ณ‡ๆบ็š„็‰ˆๆœฌใ€‚ๅฆ‚ๆžœๆ‚จๆœ‰ๆ„้ก˜่ดŠๅŠฉ็ฎ—ๅŠ›๏ผŒๆญก่ฟŽ่ˆ‡ๆˆ‘่ฏ็นซใ€‚ๆˆ‘็›ธไฟก่‹ฅ่ƒฝๅฐ‡ๆ›ดๅคšๅทฒๆบ–ๅ‚™ไฝ†ๅฐšๆœช็ดๅ…ฅ่จ“็ทด็š„ๆณ•ๅพ‹่ชžๆ–™้€ฒ่กŒๅพฎ่ชฟ๏ผŒ่ฉฒๆจกๅž‹ๅฐ‡่ƒฝ้”ๅˆฐ็น้ซ”ไธญๆ–‡ๆณ•ๅพ‹้ ˜ๅŸŸ็š„ๆœ€ไฝณ่กจ็พใ€‚

ๆŒ็ธพๆ›ดๆ–ฐ

ๆญคๆจกๅž‹ๅฆ‚ๆœ‰้€ฒไธ€ๆญฅ่ณ‡ๆบ๏ผŒๅฐ‡ๆœƒไธๅฎšๆœŸๆ›ดๆ–ฐ๏ผŒๆœ‰้—œๆจกๅž‹ๆœ€ๆ–ฐๆถˆๆฏ่ซ‹่ฆ‹ Model Update History ็ซ ็ฏ€ใ€‚

Model Card Authors

Huang Liang Hsun

Model Card Contact

Huang Liang Hsun

Framework versions

  • Transformers 4.45.2
  • Pytorch 2.4.1+cu121
  • Datasets 2.21.0
  • Tokenizers 0.20.0
Downloads last month
5
Safetensors
Model size
3.21B params
Tensor type
BF16
ยท
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support

Model tree for lianghsun/Llama-3.2-Taiwan-Legal-3B-Instruct

Finetuned
(3)
this model

Datasets used to train lianghsun/Llama-3.2-Taiwan-Legal-3B-Instruct

Spaces using lianghsun/Llama-3.2-Taiwan-Legal-3B-Instruct 8

Collection including lianghsun/Llama-3.2-Taiwan-Legal-3B-Instruct

Evaluation results