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.

See axolotl config

axolotl version: 0.13.0.dev0

base_model: Jboadu/test-model-1-pretrain
tokenizer_type: AutoTokenizer
model_type: AutoModelForCausalLM
load_in_8bit: false
load_in_4bit: false
strict: false
datasets:
- path: representation_variation_GAIA_Raw_Training_Data.jsonl
  type: completion
- path: text_chunks_GAIA_Raw_Training_Data.jsonl
  type: completion
- path: inferred_facts_GAIA_Raw_Training_Data.jsonl
  type: completion
dataset_prepared_path: last_run_prepared
output_dir: ./model-output
seed: 1337
sequence_len: 5000
sample_packing: true
pad_to_sequence_len: false
shuffle_merged_datasets: true
gradient_accumulation_steps: 75
micro_batch_size: 2
eval_batch_size: 4
num_epochs: 6
optimizer: paged_adamw_8bit
lr_scheduler: constant
learning_rate: 2.0e-05
noisy_embedding_alpha: 5
weight_decay: 0
train_on_inputs: false
group_by_length: false
bf16: true
fp16: false
tf32: false
gradient_checkpointing: true
logging_steps: 1
xformers_attention: false
flash_attention: true
chat_template: chatml
auto_resume_from_checkpoints: false
warmup_ratio: 0.1
evals_per_epoch: 1
val_set_size: 0.04
saves_per_epoch: 1
eval_sample_packing: false
save_total_limit: 2
special_tokens:
  pad_token: <unk>
use_liger_kernel: true
plugins:
- axolotl.integrations.liger.LigerPlugin
liger_rope: true
liger_rms_norm: true
liger_glu_activation: true
liger_layer_norm: true
liger_fused_linear_cross_entropy: true
sequence_length: 10000
wandb_project: test-project
wandb_entity: ''
wandb_watch: ''
wandb_run_id: ''
wandb_log_model: ''
hub_model_id: Jboadu/test-model-3-pretrain
hub_strategy: all_checkpoints

test-model-3-pretrain

This model is a fine-tuned version of Jboadu/test-model-1-pretrain on the representation_variation_GAIA_Raw_Training_Data.jsonl, the text_chunks_GAIA_Raw_Training_Data.jsonl and the inferred_facts_GAIA_Raw_Training_Data.jsonl datasets. It achieves the following results on the evaluation set:

Loss: 2.5083
Memory/max Mem Active(gib): 31.49
Memory/max Mem Allocated(gib): 31.49
Memory/device Mem Reserved(gib): 34.43

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: 2e-05
train_batch_size: 2
eval_batch_size: 4
seed: 1337
gradient_accumulation_steps: 75
total_train_batch_size: 150
optimizer: Use OptimizerNames.PAGED_ADAMW_8BIT with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: constant
lr_scheduler_warmup_ratio: 0.1
num_epochs: 6.0

Training results

Training Loss	Epoch	Step	Validation Loss	Mem Active(gib)	Mem Allocated(gib)	Mem Reserved(gib)
No log	0	0	2.5498	17.05	17.05	19.62
1.5176	1.0	1	3.4410	31.28	31.28	33.24
2.4565	2.0	2	2.7164	31.49	31.49	34.43
1.5221	3.0	3	2.6828	31.49	31.49	34.43
1.1843	4.0	4	2.5067	31.49	31.49	34.43
0.9714	5.0	5	2.5435	31.49	31.49	34.43
0.8245	6.0	6	2.5083	31.49	31.49	34.43