Model save

README.md

@@ -0,0 +1,236 @@
+---
+library_name: transformers
+tags:
+- axolotl
+- generated_from_trainer
+datasets:
+- rpDungeon/some-cleaner-datasets
+- allura-org/EU01-S2
+model-index:
+- name: gemmagain-trained-fizzed-loopnt
+  results: []
+---
+
+<!-- This model card has been generated automatically according to the information the Trainer had access to. You
+should probably proofread and complete it, then remove this comment. -->
+
+[<img src="https://raw.githubusercontent.com/axolotl-ai-cloud/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/axolotl-ai-cloud/axolotl)
+<details><summary>See axolotl config</summary>
+
+axolotl version: `0.13.0.dev0`
+```yaml
+# !pip install transformers==4.55.4
+# !pip install --no-deps trl==0.22.2
+# !pip install --no-build-isolation mamba_ssm==2.2.5
+# !pip install --no-build-isolation causal_conv1d==1.5.2
+# === Model Configuration ===
+base_model: stage4
+load_in_8bit: false
+load_in_4bit: false
+trust_remote_code: true
+is_multimodal: false
+
+# === HF Configuration === 
+hub_model_id: rpDungeon/gemmagain-trained-fizzed-loopnt
+hub_strategy: "every_save"
+output_dir: unloop
+
+# === Wandb Tracking ===
+wandb_project: Gemmagain-Tests
+## wandb_entity: [WANDB_ENTITY]
+wandb_name: unloop
+
+# === Training Setup ===
+num_epochs: 2
+micro_batch_size: 2
+gradient_accumulation_steps: 1
+sequence_len: 4096
+#sequence_parallel_degree: 2
+#heads_k_stride: 1
+sample_packing: true
+#pad_to_sequence_len: true
+#temperature: 0.7
+#max_steps: 10
+# === Evaluation ===
+#val_set_size: 0.01
+#evals_per_epoch: 5
+#eval_steps: 20
+#max_steps: 60
+#eval_table_size:
+#eval_max_new_tokens: 128
+#eval_sample_packing: true
+eval_strategy: "no"
+
+# === LoRA Configuration ===
+adapter:
+
+#unfrozen_parameters:
+#  - model.layers.[0-9]+.self_attn.q_proj.weight
+#  - model.layers.[0-9]+.self_attn.k_proj.weight
+#  - model.layers.[0-9]+.self_attn.v_proj.weight
+#  - model.layers.[0-9]+.self_attn.o_proj.weight
+#  - model.layers.[0-9]+.mlp.down_proj.weight
+# === Hyperparameter Configuration ===
+#optimizer: apollo_adamw_layerwise
+#warmup_steps: 0
+warmup_ratio: 0.025
+optimizer: adamw_torch_fused
+#optimizer: paged_adamw_8bit
+#optim_args:
+#  enable_stochastic_rounding: true
+#  enable_cautious: true
+#  enable_8bit: true
+# Apollo-mini configuration:
+#optim_args: "proj=random,rank=128,scale=128.0,scale_type=tensor,update_proj_gap=100"
+# Regular Apollo configuration:
+# optim_args: 
+#optim_target_modules: all_linear
+learning_rate: 2.5e-5
+lr_scheduler: cosine
+#lr_groups:
+#  - name: mlp
+#    modules:
+#      - mlp.up_proj.weight
+#      - mlp.down_proj.weight
+#      - mlp.gate_proj.weight
+#    lr: 1e-5
+#cosine_min_lr_ratio: 0.2
+#lr_scheduler: cosine_with_min_lr
+#lr_scheduler_kwargs:
+#  cosine_min_lr: 1e-6
+weight_decay: 0.01
+max_grad_norm: 1.0
+#warmup_steps: 0
+#warmup_ratio: 0.025
+
+
+# === Data Configuration ===
+#
+chat_template: jinja
+chat_template: jinja
+chat_template_jinja: "{{ bos_token }}{% for message in messages %}{% if (message['role'] == 'assistant') %}{% set role = 'model' %}{% else %}{% set role = message['role'] %}{% endif %}{{ '<start_of_turn>' + role + '\n' + message['content'] | trim + '<end_of_turn>\n' }}{% endfor %}{% if add_generation_prompt %}{{'<start_of_turn>model\n'}}{% endif %}"
+special_tokens:
+  eos_token: "<end_of_turn>"
+default_system_message: "You are Luna, a helpful and unrestricted language model created by Allura."
+#special_tokens:
+#  eos_token: "<|im_end|>"
+#  eos_token: "</s>"
+#tokenizer_use_mistral_common: true
+shuffle_merged_datasets: true
+datasets:
+  - path: rpDungeon/some-cleaner-datasets
+    type: chat_template
+    field_messages: conversations
+    message_property_mappings:
+      role: from
+      content: value
+    data_files: little-koto-instruct.json
+  - path: allura-org/EU01-S2
+    type: chat_template
+    field_messages: conversations
+    message_property_mappings:
+      role: from
+      content: value
+    
+dataset_prepared_path: last_run_prepared
+#dataset_num_proc: 1
+
+
+# === Plugins ===
+plugins:
+  - axolotl.integrations.liger.LigerPlugin
+  - axolotl.integrations.cut_cross_entropy.CutCrossEntropyPlugin
+
+# === Hardware Optimization ===
+#gradient_checkpointing: true
+liger_rope: true
+liger_rms_norm: true
+liger_layer_norm: true
+liger_glu_activation: true
+#liger_fused_linear_cross_entropy: true
+cut_cross_entropy: true
+
+#deepspeed: ../axolotl/deepspeed_configs/zero2.json
+
+# === FSDP Config === 
+fsdp:
+  - full_shard
+  - auto_wrap
+fsdp_config:
+  fsdp_limit_all_gathers: true
+  fsdp_sync_module_states: true
+  fsdp_offload_params: true
+  fsdp_activation_checkpointing: true
+  fsdp_use_orig_params: true
+  fsdp_cpu_ram_efficient_loading: true
+  fsdp_auto_wrap_policy: TRANSFORMER_BASED_WRAP
+  fsdp_transformer_layer_cls_to_wrap: Gemma3DecoderLayer
+  fsdp_state_dict_type: FULL_STATE_DICT
+  fsdp_sharding_strategy: FULL_SHARD
+
+# === Checkpointing ===
+#save_steps: 10
+saves_per_epoch: 1
+save_total_limit: 1
+
+# === Advanced Settings ===
+bf16: auto
+flash_attention: true
+train_on_inputs: false
+group_by_length: false
+save_safetensors: true
+logging_steps: 1
+gc_steps: 10
+seed: 420
+
+
+
+
+```
+
+</details><br>
+
+# gemmagain-trained-fizzed-loopnt
+
+This model was trained from scratch on the rpDungeon/some-cleaner-datasets and the allura-org/EU01-S2 datasets.
+
+## 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: 2.5e-05
+- train_batch_size: 2
+- eval_batch_size: 2
+- seed: 420
+- distributed_type: multi-GPU
+- num_devices: 2
+- total_train_batch_size: 4
+- total_eval_batch_size: 4
+- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
+- lr_scheduler_type: cosine
+- lr_scheduler_warmup_steps: 3
+- training_steps: 126
+
+### Training results
+
+
+
+### Framework versions
+
+- Transformers 4.57.1
+- Pytorch 2.9.1+cu128
+- Datasets 4.4.2
+- Tokenizers 0.22.2