RichardErkhov/gtonkov_-_SpecAI-gguf

Quantization made by Richard Erkhov.

Github

Discord

Request more models

SpecAI - GGUF

• Model creator: https://huggingface.co/gtonkov/
• Original model: https://huggingface.co/gtonkov/SpecAI/

Name	Quant method	Size
SpecAI.Q2_K.gguf	Q2_K	1.32GB
SpecAI.Q3_K_S.gguf	Q3_K_S	1.57GB
SpecAI.Q3_K.gguf	Q3_K	1.82GB
SpecAI.Q3_K_M.gguf	Q3_K_M	1.82GB
SpecAI.Q3_K_L.gguf	Q3_K_L	1.94GB
SpecAI.IQ4_XS.gguf	IQ4_XS	1.93GB
SpecAI.Q4_0.gguf	Q4_0	2.03GB
SpecAI.IQ4_NL.gguf	IQ4_NL	2.04GB
SpecAI.Q4_K_S.gguf	Q4_K_S	2.04GB
SpecAI.Q4_K.gguf	Q4_K	2.23GB
SpecAI.Q4_K_M.gguf	Q4_K_M	2.23GB
SpecAI.Q4_1.gguf	Q4_1	2.24GB
SpecAI.Q5_0.gguf	Q5_0	2.46GB
SpecAI.Q5_K_S.gguf	Q5_K_S	2.46GB
SpecAI.Q5_K.gguf	Q5_K	2.62GB
SpecAI.Q5_K_M.gguf	Q5_K_M	2.62GB
SpecAI.Q5_1.gguf	Q5_1	2.68GB
SpecAI.Q6_K.gguf	Q6_K	2.92GB
SpecAI.Q8_0.gguf	Q8_0	3.78GB

Original model description:

library_name: transformers license: other base_model: microsoft/Phi-3.5-mini-instruct tags: - llama-factory - full - generated_from_trainer model-index: - name: SpecAI results: []

SpecAI

This model is a fine-tuned version of microsoft/Phi-3.5-mini-instruct on the api_eval_dataset dataset.

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 3e-06
• train_batch_size: 4
• eval_batch_size: 8
• seed: 42
• gradient_accumulation_steps: 4
• total_train_batch_size: 16
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: cosine
• num_epochs: 32.0

Training results

Framework versions

• Transformers 4.44.2
• Pytorch 2.4.1+cu121
• Datasets 2.21.0
• Tokenizers 0.19.1