You can deploy this model and start earning money today!

Fischerboot/
phi3-mini-28k-inst-adapter-m

Name: phi3-mini-28k-inst-adapter-m
Availability: InStock

peft

phi3

generated_from_trainer

custom_code

microsoft/Phi-3-mini-128k-instruct

adapter

mit

Views

No views yet

See axolotl config

axolotl version: 0.4.1

yaml

base_model: microsoft/Phi-3-mini-128k-instruct
trust_remote_code: true
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer
chat_template: phi_3

load_in_8bit: false
load_in_4bit: false
strict: false

datasets:
  - path: Fischerboot/freedom-rp-alpaca-shortend
    type: alpaca:phi
  - path: Fischerboot/mongotom-40k-alpaca
    type: alpaca:phi
  - path: Fischerboot/DAN-alpaca
    type: alpaca:phi

dataset_prepared_path:
val_set_size: 0.01
output_dir: ./out/yuh

sequence_len: 1024
sample_packing: true
pad_to_sequence_len: true

adapter: lora
lora_model_dir:
lora_r: 64
lora_alpha: 32
lora_dropout: 0.05
lora_target_linear: true
lora_fan_in_fan_out:

gradient_accumulation_steps: 1
micro_batch_size: 2
num_epochs: 4
optimizer: adamw_torch
adam_beta2: 0.95
adam_epsilon: 0.00001
max_grad_norm: 1.0
lr_scheduler: cosine
learning_rate: 5.0e-6

train_on_inputs: false
group_by_length: false
bf16: auto

gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention:
flash_attention: true

warmup_steps: 10
evals_per_epoch: 1
saves_per_epoch: 1
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:

out/yuh

This model is a fine-tuned version of microsoft/Phi-3-mini-128k-instruct on the None dataset. It achieves the following results on the evaluation set:

Loss: 4.0955

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: 5e-06
train_batch_size: 2
eval_batch_size: 2
seed: 42
optimizer: Adam with betas=(0.9,0.95) and epsilon=1e-05
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 10
num_epochs: 4

Training results

Training Loss	Epoch	Step	Validation Loss
4.296	0.0007	1	4.7549
4.3774	1.0	1521	4.1032
3.5409	1.9855	3042	4.0949
3.8041	2.9711	4563	4.0953
3.9558	3.9560	6084	4.0955

Framework versions

PEFT 0.11.1
Transformers 4.42.3
Pytorch 2.1.2+cu118
Datasets 2.19.1
Tokenizers 0.19.1

Fischerboot/phi3-mini-28k-inst-adapter-m

out/yuh

Model description

Intended uses & limitations

Training and evaluation data

Training procedure

Training hyperparameters

Training results

Framework versions

Fischerboot/
phi3-mini-28k-inst-adapter-m