See axolotl config
axolotl version: 0.4.1
base_model: meta-llama/Meta-Llama-3-70B
model_type: LlamaForCausalLM
tokenizer_type: AutoTokenizer
load_in_8bit: false
load_in_4bit: true
strict: false
datasets:
- path: awilliamson/qbank_conversations
type: chat_template
chat_template: llama3
field_messages: conversations
message_field_role: from
message_field_content: value
roles:
system:
- system
user:
- user
assistant:
- assistant
chat_template: llama3
adapter: qlora
lora_r: 32
lora_alpha: 16
lora_modules_to_save: [embed_tokens, lm_head]
lora_dropout: 0.05
lora_target_linear: true
dataset_prepared_path: last_run_prepared
val_set_size: 0.05
output_dir: ./output/llama3-70b
sequence_len: 4096
sample_packing: false
pad_to_sequence_len: true
wandb_project: llama-70b
wandb_watch:
wandb_run_id:
wandb_log_model:
gradient_accumulation_steps: 4
micro_batch_size: 1
num_epochs: 3
optimizer: adamw_torch
lr_scheduler: cosine
learning_rate: 1e-5
train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: false
gradient_checkpointing: true
gradient_checkpointing_kwargs:
use_reentrant: false
early_stopping_patience:
resume_from_checkpoint:
logging_steps: 1
xformers_attention:
flash_attention: true
warmup_steps: 15
evals_per_epoch: 5
eval_table_size:
saves_per_epoch: 1
save_total_limit: 10
save_steps:
debug:
weight_decay: 0.00
fsdp:
- full_shard
- auto_wrap
fsdp_config:
fsdp_limit_all_gathers: true
fsdp_sync_module_states: true
fsdp_offload_params: true
fsdp_use_orig_params: false
fsdp_cpu_ram_efficient_loading: true
fsdp_auto_wrap_policy: TRANSFORMER_BASED_WRAP
fsdp_transformer_layer_cls_to_wrap: LlamaDecoderLayer
fsdp_state_dict_type: FULL_STATE_DICT
fsdp_sharding_strategy: FULL_SHARD
special_tokens:
pad_token: "<|end_of_text|>"
output/llama3-70b
This model is a fine-tuned version of meta-llama/Meta-Llama-3-70B on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.3901
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: 1e-05
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- num_devices: 4
- gradient_accumulation_steps: 4
- total_train_batch_size: 16
- total_eval_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 15
- num_epochs: 3
Training results
Training Loss | Epoch | Step | Validation Loss |
---|---|---|---|
2.3783 | 0.0388 | 1 | 2.8294 |
1.2438 | 0.1942 | 5 | 1.4718 |
1.1973 | 0.3883 | 10 | 1.4697 |
1.0995 | 0.5825 | 15 | 1.4572 |
1.181 | 0.7767 | 20 | 1.4470 |
1.1298 | 0.9709 | 25 | 1.4350 |
0.9058 | 1.1650 | 30 | 1.4232 |
0.8712 | 1.3592 | 35 | 1.4126 |
0.8735 | 1.5534 | 40 | 1.4051 |
0.8975 | 1.7476 | 45 | 1.4024 |
0.929 | 1.9417 | 50 | 1.3951 |
0.9181 | 2.1359 | 55 | 1.3923 |
0.9171 | 2.3301 | 60 | 1.3917 |
0.9111 | 2.5243 | 65 | 1.3907 |
0.9676 | 2.7184 | 70 | 1.3904 |
0.8497 | 2.9126 | 75 | 1.3901 |
Framework versions
- PEFT 0.11.1
- Transformers 4.41.1
- Pytorch 2.1.2+cu121
- Datasets 2.19.1
- Tokenizers 0.19.1
- Downloads last month
- 2
Model tree for awilliamson/qbank
Base model
meta-llama/Meta-Llama-3-70B
Adapter
this model