Edit model card

Built with Axolotl

See axolotl config

axolotl version: 0.4.1

base_model: meta-llama/Llama-3.2-3B
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer

plugins:
  - axolotl.integrations.liger.LigerPlugin
liger_rope: true
liger_rms_norm: true
liger_swiglu: true
liger_fused_linear_cross_entropy: true

load_in_8bit: false
load_in_4bit: false
strict: false

datasets:
#  - path: anthracite-core/c2_logs_32k_mistral-v3_v1.2
#    type: sharegpt
#    conversation: chatml
  - path: ./datasets/c2_deduped_32k_mistral-v3_tok_deanon_dsclean_1.2.jsonl
    type: sharegpt
    conversation: chatml
#  - path: anthracite-org/kalo-opus-instruct-22k-no-refusal
#    type: sharegpt
#    conversation: chatml
  - path: ./datasets/opus-instruct-22k-no_refusals.jsonl
    type: sharegpt
    conversation: chatml
#  - path: lodrick-the-lafted/kalo-opus-instruct-3k-filtered
#    type: sharegpt
#    conversation: chatml
  - path: ./datasets/kalo-3k-filtered.jsonl
    type: sharegpt
    conversation: chatml
#  - path: anthracite-org/nopm_claude_writing_fixed
#    type: sharegpt
#    conversation: chatml
  - path: ./datasets/claudewritingNopm.jsonl
    type: sharegpt
    conversation: chatml
#  - path: anthracite-org/kalo_opus_misc_240827
#    type: sharegpt
#    conversation: chatml
  - path: ./datasets/kalo_opus_misc_240827.jsonl
    type: sharegpt
    conversation: chatml
#  - path: anthracite-org/kalo_misc_part2
#    type: sharegpt
#    conversation: chatml
  - path: ./datasets/kalo_misc_part2.jsonl
    type: sharegpt
    conversation: chatml
#  - path: NewEden/Claude-Instruct-5K
#    type: sharegpt
#    conversation: chatml
  - path: ./datasets/5k.jsonl
    type: sharegpt
    conversation: chatml

#chat_template: chatml
shuffle_merged_datasets: true
#default_system_message: "You are an assistant that responds to the user."
dataset_prepared_path: ./magnum-22b-data
val_set_size: 0.0
output_dir: ./22b-fft-out

sequence_len: 16000
sample_packing: true
pad_to_sequence_len: true


wandb_project: 3bmagnum
wandb_entity:
wandb_watch:
wandb_name: 3magnum
wandb_log_model:

gradient_accumulation_steps: 32
micro_batch_size: 1
num_epochs: 2
optimizer: adamw_bnb_8bit
lr_scheduler: cosine
learning_rate: 0.000005

train_on_inputs: false
group_by_length: false
bf16: auto
fp16:
tf32: false

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

warmup_steps: 40
evals_per_epoch:
eval_table_size:
eval_max_new_tokens:
saves_per_epoch: 2
debug:
#deepspeed: /workspace/axolotl/deepspeed_configs/zero3_bf16.json
deepspeed: /workspace/axolotl/deepspeed_configs/zero2.json
weight_decay: 0.01
fsdp:
fsdp_config:
special_tokens:
  pad_token: <|finetune_right_pad_id|>

22b-fft-out

This model is a fine-tuned version of meta-llama/Llama-3.2-3B on the None 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: 5e-06
  • train_batch_size: 1
  • eval_batch_size: 1
  • seed: 42
  • distributed_type: multi-GPU
  • num_devices: 2
  • gradient_accumulation_steps: 32
  • total_train_batch_size: 64
  • total_eval_batch_size: 2
  • optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
  • lr_scheduler_type: cosine
  • lr_scheduler_warmup_steps: 40
  • num_epochs: 2

Training results

Framework versions

  • Transformers 4.45.0.dev0
  • Pytorch 2.4.0+cu121
  • Datasets 2.19.1
  • Tokenizers 0.19.1
Downloads last month
14
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Model tree for Edens-Gate/Holland-3b

Finetuned
(24)
this model