Upload folder using huggingface_hub

README.md

@@ -0,0 +1,136 @@
+---
+license: apache-2.0
+base_model: allenai/OLMo-7B
+tags:
+- generated_from_trainer
+model-index:
+- name: ollama-7B-Tinybook-epochs-1-lr-0002
+  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/OpenAccess-AI-Collective/axolotl/main/image/axolotl-badge-web.png" alt="Built with Axolotl" width="200" height="32"/>](https://github.com/OpenAccess-AI-Collective/axolotl)
+<details><summary>See axolotl config</summary>
+
+axolotl version: `0.4.0`
+```yaml
+base_model: allenai/OLMo-7B
+tokenizer_type: AutoTokenizer
+model_type: AutoModelForCausalLM
+trust_remote_code: true
+
+load_in_8bit: false
+load_in_4bit: false
+strict: false
+
+datasets:
+  - path: utrgvseniorproject/Tinybook
+    type: completion
+dataset_prepared_path: /home/josegomez15/med-llm/last_run_prepared
+val_set_size: 0.05
+output_dir: ./ollama-7B-Tinybook-epochs-1-lr-0002
+
+sequence_len: 4096
+sample_packing: false
+pad_to_sequence_len: true
+
+wandb_project: olmo-7B-Tinybook
+wandb_entity: utrgvmedai
+wandb_watch:
+wandb_name: olmo-7B-Tinybook-epochs-1-lr-0002
+wandb_log_model:
+
+gradient_accumulation_steps: 1
+micro_batch_size: 1
+num_epochs: 1
+optimizer: adamw_bnb_8bit
+lr_scheduler: cosine
+learning_rate: 0.0002
+
+train_on_inputs: True # make sure you have this on True
+group_by_length: false
+bf16: auto
+fp16:
+tf32: false
+
+gradient_checkpointing: false #olmo doesn't support
+early_stopping_patience:
+resume_from_checkpoint:
+local_rank:
+logging_steps: 1
+xformers_attention:
+flash_attention: true
+flash_attn_cross_entropy: false
+flash_attn_rms_norm: true
+flash_attn_fuse_qkv: false
+flash_attn_fuse_mlp: true
+
+warmup_steps: 100
+evals_per_epoch: 4
+eval_table_size:
+eval_sample_packing:
+saves_per_epoch: 1
+debug:
+deepspeed: /home/josegomez15/axolotl/deepspeed_configs/zero2.json
+weight_decay: 0.1
+fsdp:
+fsdp_config:
+special_tokens:
+
+```
+
+</details><br>
+
+# ollama-7B-Tinybook-epochs-1-lr-0002
+
+This model is a fine-tuned version of [allenai/OLMo-7B](https://huggingface.co/allenai/OLMo-7B) on the None dataset.
+It achieves the following results on the evaluation set:
+- Loss: 2.3906
+
+## 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: 0.0002
+- train_batch_size: 1
+- eval_batch_size: 1
+- seed: 42
+- distributed_type: multi-GPU
+- num_devices: 4
+- total_train_batch_size: 4
+- 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: 100
+- num_epochs: 1
+
+### Training results
+
+| Training Loss | Epoch | Step | Validation Loss |
+|:-------------:|:-----:|:----:|:---------------:|
+| 4.3047        | 0.33  | 1    | 2.4062          |
+| 4.0859        | 0.67  | 2    | 2.3906          |
+| 3.9805        | 1.0   | 3    | 2.3906          |
+
+
+### Framework versions
+
+- Transformers 4.38.0
+- Pytorch 2.0.1+cu117
+- Datasets 2.17.0
+- Tokenizers 0.15.0

checkpoint-3/config.json

@@ -0,0 +1,54 @@
+{
+  "_name_or_path": "allenai/OLMo-7B",
+  "activation_type": "swiglu",
+  "alibi": false,
+  "alibi_bias_max": 8.0,
+  "architectures": [
+    "OLMoForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "attention_layer_norm": false,
+  "attention_layer_norm_with_affine": false,
+  "auto_map": {
+    "AutoConfig": "allenai/OLMo-7B--configuration_olmo.OLMoConfig",
+    "AutoModelForCausalLM": "allenai/OLMo-7B--modeling_olmo.OLMoForCausalLM",
+    "AutoTokenizer": [
+      "allenai/OLMo-7B--tokenization_olmo_fast.OLMoTokenizerFast",
+      "allenai/OLMo-7B--tokenization_olmo_fast.OLMoTokenizerFast"
+    ]
+  },
+  "bias_for_layer_norm": false,
+  "block_group_size": 1,
+  "block_type": "sequential",
+  "clip_qkv": null,
+  "d_model": 4096,
+  "embedding_dropout": 0.0,
+  "embedding_size": 50304,
+  "eos_token_id": 50279,
+  "flash_attention": true,
+  "include_bias": false,
+  "init_cutoff_factor": null,
+  "init_device": "meta",
+  "init_fn": "mitchell",
+  "init_std": 0.02,
+  "layer_norm_type": "default",
+  "layer_norm_with_affine": false,
+  "max_sequence_length": 4096,
+  "mlp_hidden_size": 22016,
+  "mlp_ratio": 4,
+  "model_type": "olmo",
+  "multi_query_attention": false,
+  "n_heads": 32,
+  "n_layers": 32,
+  "pad_token_id": 1,
+  "precision": "amp_bf16",
+  "residual_dropout": 0.0,
+  "rope": true,
+  "rope_full_precision": true,
+  "scale_logits": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.38.0",
+  "use_cache": false,
+  "vocab_size": 50280,
+  "weight_tying": false
+}

checkpoint-3/generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "do_sample": true,
+  "eos_token_id": 50279,
+  "pad_token_id": 1,
+  "transformers_version": "4.38.0"
+}

checkpoint-3/global_step3/bf16_zero_pp_rank_0_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63a7b7b9b2801f7c079e6aa8345501b0954bfc1633ddeb4d3e0d639ba0bc4e1b
+size 20664292727

checkpoint-3/global_step3/bf16_zero_pp_rank_1_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d48b33c2be4c651e12f67a10457c38350096d5ffebbe570a791418c5fd00ceb
+size 20664292919

checkpoint-3/global_step3/bf16_zero_pp_rank_2_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dd8082b85ba824e6a6527da02b9ccb73de71ff37de2be4731a05d13c3a2d6080
+size 20664292983

checkpoint-3/global_step3/bf16_zero_pp_rank_3_mp_rank_00_optim_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:10415893bf48db52c11cacb574381cf4071cc2cb281b968832e6bd7a35208d0d
+size 20664292663

checkpoint-3/global_step3/mp_rank_00_model_states.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:291003696ba5e9aea177be16ea2c80480fba140495d55c748b3905376cfdb5f3
+size 13776240835

checkpoint-3/latest

@@ -0,0 +1 @@
+global_step3

checkpoint-3/model-00001-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:40ac7bae36ad0f9d5cf483940b2baa65253faf0449b69e426605bca31f62d30a
+size 4988081728

checkpoint-3/model-00002-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:493c0263ab7dff25d1efd94fbf65cb18df5addf1d1e80db79401f5eb64072660
+size 4957673320

checkpoint-3/model-00003-of-00003.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:66cc675348c6b927bfab49d97833388dc741b612ae60da36890d296859575737
+size 3830452280

checkpoint-3/model.safetensors.index.json

@@ -0,0 +1,137 @@
+{
+  "metadata": {
+    "total_size": 13776191488
+  },
+  "weight_map": {
+    "model.transformer.blocks.0.att_proj.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.0.attn_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.0.ff_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.0.ff_proj.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.1.att_proj.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.1.attn_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.1.ff_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.1.ff_proj.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.10.att_proj.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.10.attn_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.10.ff_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.10.ff_proj.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.11.att_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.11.attn_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.11.ff_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.11.ff_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.12.att_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.12.attn_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.12.ff_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.12.ff_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.13.att_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.13.attn_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.13.ff_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.13.ff_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.14.att_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.14.attn_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.14.ff_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.14.ff_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.15.att_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.15.attn_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.15.ff_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.15.ff_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.16.att_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.16.attn_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.16.ff_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.16.ff_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.17.att_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.17.attn_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.17.ff_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.17.ff_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.18.att_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.18.attn_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.18.ff_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.18.ff_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.19.att_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.19.attn_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.19.ff_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.19.ff_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.2.att_proj.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.2.attn_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.2.ff_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.2.ff_proj.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.20.att_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.20.attn_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.20.ff_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.20.ff_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.21.att_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.21.attn_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.21.ff_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.21.ff_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.22.att_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.22.attn_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.22.ff_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.22.ff_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.23.att_proj.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.23.attn_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.23.ff_out.weight": "model-00002-of-00003.safetensors",
+    "model.transformer.blocks.23.ff_proj.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.24.att_proj.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.24.attn_out.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.24.ff_out.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.24.ff_proj.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.25.att_proj.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.25.attn_out.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.25.ff_out.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.25.ff_proj.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.26.att_proj.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.26.attn_out.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.26.ff_out.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.26.ff_proj.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.27.att_proj.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.27.attn_out.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.27.ff_out.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.27.ff_proj.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.28.att_proj.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.28.attn_out.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.28.ff_out.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.28.ff_proj.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.29.att_proj.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.29.attn_out.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.29.ff_out.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.29.ff_proj.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.3.att_proj.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.3.attn_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.3.ff_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.3.ff_proj.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.30.att_proj.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.30.attn_out.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.30.ff_out.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.30.ff_proj.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.31.att_proj.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.31.attn_out.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.31.ff_out.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.31.ff_proj.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.blocks.4.att_proj.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.4.attn_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.4.ff_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.4.ff_proj.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.5.att_proj.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.5.attn_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.5.ff_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.5.ff_proj.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.6.att_proj.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.6.attn_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.6.ff_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.6.ff_proj.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.7.att_proj.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.7.attn_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.7.ff_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.7.ff_proj.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.8.att_proj.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.8.attn_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.8.ff_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.8.ff_proj.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.9.att_proj.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.9.attn_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.9.ff_out.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.blocks.9.ff_proj.weight": "model-00001-of-00003.safetensors",
+    "model.transformer.ff_out.weight": "model-00003-of-00003.safetensors",
+    "model.transformer.wte.weight": "model-00001-of-00003.safetensors"
+  }
+}

checkpoint-3/rng_state_0.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9dffec05795b7d3ced5821c10057fe393a503834b279b96f3870374fcca567f6
+size 17655

checkpoint-3/rng_state_1.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da391f6fcee77d45d8316ae2541c4354a1d2bc68d26de374ff9e6cd1d2442841
+size 17655

checkpoint-3/rng_state_2.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c370271af94095d17cfd7e55ac9fa5025538e19c56b84380de7aa37e0aa82e4
+size 17655

checkpoint-3/rng_state_3.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:774c413b2a6747fe2e8a7ff31b4415735fbaa15708f26f8c71826be445fb4476
+size 17655

checkpoint-3/scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8882a55889f61ac16c8f5bc7fa29711e514f512f5598fd5d0dc1752fa7d6f7f7
+size 627

checkpoint-3/trainer_state.json

@@ -0,0 +1,66 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 1.0,
+  "eval_steps": 1,
+  "global_step": 3,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.33,
+      "grad_norm": 320.38479062293203,
+      "learning_rate": 2.0000000000000003e-06,
+      "loss": 4.3047,
+      "step": 1
+    },
+    {
+      "epoch": 0.33,
+      "eval_loss": 2.40625,
+      "eval_runtime": 0.5413,
+      "eval_samples_per_second": 1.848,
+      "eval_steps_per_second": 1.848,
+      "step": 1
+    },
+    {
+      "epoch": 0.67,
+      "grad_norm": 208.84861069085426,
+      "learning_rate": 4.000000000000001e-06,
+      "loss": 4.0859,
+      "step": 2
+    },
+    {
+      "epoch": 0.67,
+      "eval_loss": 2.390625,
+      "eval_runtime": 0.536,
+      "eval_samples_per_second": 1.866,
+      "eval_steps_per_second": 1.866,
+      "step": 2
+    },
+    {
+      "epoch": 1.0,
+      "grad_norm": 152.65601650442736,
+      "learning_rate": 6e-06,
+      "loss": 3.9805,
+      "step": 3
+    },
+    {
+      "epoch": 1.0,
+      "eval_loss": 2.390625,
+      "eval_runtime": 0.5404,
+      "eval_samples_per_second": 1.851,
+      "eval_steps_per_second": 1.851,
+      "step": 3
+    }
+  ],
+  "logging_steps": 1,
+  "max_steps": 3,
+  "num_input_tokens_seen": 0,
+  "num_train_epochs": 1,
+  "save_steps": 500,
+  "total_flos": 1970616894750720.0,
+  "train_batch_size": 1,
+  "trial_name": null,
+  "trial_params": null
+}

checkpoint-3/training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:52a72ebc0ab2a41065c71c4d03ae7640fa417b58b636d8ec5a15d65b0f3d9cae
+size 6395

checkpoint-3/zero_to_fp32.py

@@ -0,0 +1,592 @@
+#!/usr/bin/env python
+
+# Copyright (c) Microsoft Corporation.
+# SPDX-License-Identifier: Apache-2.0
+
+# DeepSpeed Team
+
+# This script extracts fp32 consolidated weights from a zero 1, 2 and 3 DeepSpeed checkpoints. It gets
+# copied into the top level checkpoint dir, so the user can easily do the conversion at any point in
+# the future. Once extracted, the weights don't require DeepSpeed and can be used in any
+# application.
+#
+# example: python zero_to_fp32.py . pytorch_model.bin
+
+import argparse
+import torch
+import glob
+import math
+import os
+import re
+from collections import OrderedDict
+from dataclasses import dataclass
+
+# while this script doesn't use deepspeed to recover data, since the checkpoints are pickled with
+# DeepSpeed data structures it has to be available in the current python environment.
+from deepspeed.utils import logger
+from deepspeed.checkpoint.constants import (DS_VERSION, OPTIMIZER_STATE_DICT, SINGLE_PARTITION_OF_FP32_GROUPS,
+                                            FP32_FLAT_GROUPS, ZERO_STAGE, PARTITION_COUNT, PARAM_SHAPES, BUFFER_NAMES,
+                                            FROZEN_PARAM_SHAPES, FROZEN_PARAM_FRAGMENTS)
+
+
+@dataclass
+class zero_model_state:
+    buffers: dict()
+    param_shapes: dict()
+    shared_params: list
+    ds_version: int
+    frozen_param_shapes: dict()
+    frozen_param_fragments: dict()
+
+
+debug = 0
+
+# load to cpu
+device = torch.device('cpu')
+
+
+def atoi(text):
+    return int(text) if text.isdigit() else text
+
+
+def natural_keys(text):
+    '''
+    alist.sort(key=natural_keys) sorts in human order
+    http://nedbatchelder.com/blog/200712/human_sorting.html
+    (See Toothy's implementation in the comments)
+    '''
+    return [atoi(c) for c in re.split(r'(\d+)', text)]
+
+
+def get_model_state_file(checkpoint_dir, zero_stage):
+    if not os.path.isdir(checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{checkpoint_dir}' doesn't exist")
+
+    # there should be only one file
+    if zero_stage <= 2:
+        file = os.path.join(checkpoint_dir, "mp_rank_00_model_states.pt")
+    elif zero_stage == 3:
+        file = os.path.join(checkpoint_dir, "zero_pp_rank_0_mp_rank_00_model_states.pt")
+
+    if not os.path.exists(file):
+        raise FileNotFoundError(f"can't find model states file at '{file}'")
+
+    return file
+
+
+def get_checkpoint_files(checkpoint_dir, glob_pattern):
+    # XXX: need to test that this simple glob rule works for multi-node setup too
+    ckpt_files = sorted(glob.glob(os.path.join(checkpoint_dir, glob_pattern)), key=natural_keys)
+
+    if len(ckpt_files) == 0:
+        raise FileNotFoundError(f"can't find {glob_pattern} files in directory '{checkpoint_dir}'")
+
+    return ckpt_files
+
+
+def get_optim_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_optim_states.pt")
+
+
+def get_model_state_files(checkpoint_dir):
+    return get_checkpoint_files(checkpoint_dir, "*_model_states.pt")
+
+
+def parse_model_states(files):
+    zero_model_states = []
+    for file in files:
+        state_dict = torch.load(file, map_location=device)
+
+        if BUFFER_NAMES not in state_dict:
+            raise ValueError(f"{file} is not a model state checkpoint")
+        buffer_names = state_dict[BUFFER_NAMES]
+        if debug:
+            print("Found buffers:", buffer_names)
+
+        # recover just the buffers while restoring them to fp32 if they were saved in fp16
+        buffers = {k: v.float() for k, v in state_dict["module"].items() if k in buffer_names}
+        param_shapes = state_dict[PARAM_SHAPES]
+
+        # collect parameters that are included in param_shapes
+        param_names = []
+        for s in param_shapes:
+            for name in s.keys():
+                param_names.append(name)
+
+        # update with frozen parameters
+        frozen_param_shapes = state_dict.get(FROZEN_PARAM_SHAPES, None)
+        if frozen_param_shapes is not None:
+            if debug:
+                print(f"Found frozen_param_shapes: {frozen_param_shapes}")
+            param_names += list(frozen_param_shapes.keys())
+
+        # handle shared params
+        shared_params = [[k, v] for k, v in state_dict["shared_params"].items()]
+
+        ds_version = state_dict.get(DS_VERSION, None)
+
+        frozen_param_fragments = state_dict.get(FROZEN_PARAM_FRAGMENTS, None)
+
+        z_model_state = zero_model_state(buffers=buffers,
+                                         param_shapes=param_shapes,
+                                         shared_params=shared_params,
+                                         ds_version=ds_version,
+                                         frozen_param_shapes=frozen_param_shapes,
+                                         frozen_param_fragments=frozen_param_fragments)
+        zero_model_states.append(z_model_state)
+
+    return zero_model_states
+
+
+def parse_optim_states(files, ds_checkpoint_dir):
+
+    total_files = len(files)
+    state_dicts = []
+    for f in files:
+        state_dict = torch.load(f, map_location=device)
+        # immediately discard the potentially huge 2 optimizer states as we only care for fp32 master weights
+        # and also handle the case where it was already removed by another helper script
+        state_dict["optimizer_state_dict"].pop("optimizer_state_dict", None)
+        state_dicts.append(state_dict)
+
+    if not ZERO_STAGE in state_dicts[0][OPTIMIZER_STATE_DICT]:
+        raise ValueError(f"{files[0]} is not a zero checkpoint")
+    zero_stage = state_dicts[0][OPTIMIZER_STATE_DICT][ZERO_STAGE]
+    world_size = state_dicts[0][OPTIMIZER_STATE_DICT][PARTITION_COUNT]
+
+    # For ZeRO-2 each param group can have different partition_count as data parallelism for expert
+    # parameters can be different from data parallelism for non-expert parameters. So we can just
+    # use the max of the partition_count to get the dp world_size.
+
+    if type(world_size) is list:
+        world_size = max(world_size)
+
+    if world_size != total_files:
+        raise ValueError(
+            f"Expected {world_size} of '*_optim_states.pt' under '{ds_checkpoint_dir}' but found {total_files} files. "
+            "Possibly due to an overwrite of an old checkpoint, or a checkpoint didn't get saved by one or more processes."
+        )
+
+    # the groups are named differently in each stage
+    if zero_stage <= 2:
+        fp32_groups_key = SINGLE_PARTITION_OF_FP32_GROUPS
+    elif zero_stage == 3:
+        fp32_groups_key = FP32_FLAT_GROUPS
+    else:
+        raise ValueError(f"unknown zero stage {zero_stage}")
+
+    if zero_stage <= 2:
+        fp32_flat_groups = [state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key] for i in range(len(state_dicts))]
+    elif zero_stage == 3:
+        # if there is more than one param group, there will be multiple flattened tensors - one
+        # flattened tensor per group - for simplicity merge them into a single tensor
+        #
+        # XXX: could make the script more memory efficient for when there are multiple groups - it
+        # will require matching the sub-lists of param_shapes for each param group flattened tensor
+
+        fp32_flat_groups = [
+            torch.cat(state_dicts[i][OPTIMIZER_STATE_DICT][fp32_groups_key], 0) for i in range(len(state_dicts))
+        ]
+
+    return zero_stage, world_size, fp32_flat_groups
+
+
+def _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir):
+    """
+    Returns fp32 state_dict reconstructed from ds checkpoint
+
+    Args:
+        - ``ds_checkpoint_dir``: path to the deepspeed checkpoint folder (where the optimizer files are)
+
+    """
+    print(f"Processing zero checkpoint '{ds_checkpoint_dir}'")
+
+    optim_files = get_optim_files(ds_checkpoint_dir)
+    zero_stage, world_size, fp32_flat_groups = parse_optim_states(optim_files, ds_checkpoint_dir)
+    print(f"Detected checkpoint of type zero stage {zero_stage}, world_size: {world_size}")
+
+    model_files = get_model_state_files(ds_checkpoint_dir)
+
+    zero_model_states = parse_model_states(model_files)
+    print(f'Parsing checkpoint created by deepspeed=={zero_model_states[0].ds_version}')
+
+    if zero_stage <= 2:
+        return _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states)
+    elif zero_stage == 3:
+        return _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states)
+
+
+def _zero2_merge_frozen_params(state_dict, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+    frozen_param_fragments = zero_model_states[0].frozen_param_fragments
+
+    if debug:
+        num_elem = sum(s.numel() for s in frozen_param_shapes.values())
+        print(f'rank 0: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in frozen_param_fragments.values()])
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        state_dict[name] = frozen_param_fragments[name]
+
+        if debug:
+            print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _has_callable(obj, fn):
+    attr = getattr(obj, fn, None)
+    return callable(attr)
+
+
+def _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+
+    # Reconstruction protocol:
+    #
+    # XXX: document this
+
+    if debug:
+        for i in range(world_size):
+            for j in range(len(fp32_flat_groups[0])):
+                print(f"{FP32_FLAT_GROUPS}[{i}][{j}].shape={fp32_flat_groups[i][j].shape}")
+
+    # XXX: memory usage doubles here (zero2)
+    num_param_groups = len(fp32_flat_groups[0])
+    merged_single_partition_of_fp32_groups = []
+    for i in range(num_param_groups):
+        merged_partitions = [sd[i] for sd in fp32_flat_groups]
+        full_single_fp32_vector = torch.cat(merged_partitions, 0)
+        merged_single_partition_of_fp32_groups.append(full_single_fp32_vector)
+    avail_numel = sum(
+        [full_single_fp32_vector.numel() for full_single_fp32_vector in merged_single_partition_of_fp32_groups])
+
+    if debug:
+        wanted_params = sum([len(shapes) for shapes in param_shapes])
+        wanted_numel = sum([sum(shape.numel() for shape in shapes.values()) for shapes in param_shapes])
+        # not asserting if there is a mismatch due to possible padding
+        print(f"Have {avail_numel} numels to process.")
+        print(f"Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    total_numel = 0
+    total_params = 0
+    for shapes, full_single_fp32_vector in zip(param_shapes, merged_single_partition_of_fp32_groups):
+        offset = 0
+        avail_numel = full_single_fp32_vector.numel()
+        for name, shape in shapes.items():
+
+            unpartitioned_numel = shape.numel() if _has_callable(shape, 'numel') else math.prod(shape)
+            total_numel += unpartitioned_numel
+            total_params += 1
+
+            if debug:
+                print(f"{name} full shape: {shape} unpartitioned numel {unpartitioned_numel} ")
+            state_dict[name] = full_single_fp32_vector.narrow(0, offset, unpartitioned_numel).view(shape)
+            offset += unpartitioned_numel
+
+        # Z2 started to align to 2*world_size to improve nccl performance. Therefore both offset and
+        # avail_numel can differ by anywhere between 0..2*world_size. Due to two unrelated complex
+        # paddings performed in the code it's almost impossible to predict the exact numbers w/o the
+        # live optimizer object, so we are checking that the numbers are within the right range
+        align_to = 2 * world_size
+
+        def zero2_align(x):
+            return align_to * math.ceil(x / align_to)
+
+        if debug:
+            print(f"original offset={offset}, avail_numel={avail_numel}")
+
+        offset = zero2_align(offset)
+        avail_numel = zero2_align(avail_numel)
+
+        if debug:
+            print(f"aligned  offset={offset}, avail_numel={avail_numel}")
+
+        # Sanity check
+        if offset != avail_numel:
+            raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero2_checkpoint(world_size, fp32_flat_groups, zero_model_states):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    _zero2_merge_frozen_params(state_dict, zero_model_states)
+
+    _zero2_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def zero3_partitioned_param_info(unpartitioned_numel, world_size):
+    remainder = unpartitioned_numel % world_size
+    padding_numel = (world_size - remainder) if remainder else 0
+    partitioned_numel = math.ceil(unpartitioned_numel / world_size)
+    return partitioned_numel, padding_numel
+
+
+def _zero3_merge_frozen_params(state_dict, world_size, zero_model_states):
+    if zero_model_states[0].frozen_param_shapes is None or len(zero_model_states[0].frozen_param_shapes) == 0:
+        return
+
+    if debug:
+        for i in range(world_size):
+            num_elem = sum(s.numel() for s in zero_model_states[i].frozen_param_fragments.values())
+            print(f'rank {i}: {FROZEN_PARAM_SHAPES}.numel = {num_elem}')
+
+        frozen_param_shapes = zero_model_states[0].frozen_param_shapes
+        wanted_params = len(frozen_param_shapes)
+        wanted_numel = sum(s.numel() for s in frozen_param_shapes.values())
+        avail_numel = sum([p.numel() for p in zero_model_states[0].frozen_param_fragments.values()]) * world_size
+        print(f'Frozen params: Have {avail_numel} numels to process.')
+        print(f'Frozen params: Need {wanted_numel} numels in {wanted_params} params')
+
+    total_params = 0
+    total_numel = 0
+    for name, shape in zero_model_states[0].frozen_param_shapes.items():
+        total_params += 1
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+
+        param_frags = tuple(model_state.frozen_param_fragments[name] for model_state in zero_model_states)
+        state_dict[name] = torch.cat(param_frags, 0).narrow(0, 0, unpartitioned_numel).view(shape)
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Frozen params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+    print(f"Reconstructed Frozen fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states):
+    param_shapes = zero_model_states[0].param_shapes
+    avail_numel = fp32_flat_groups[0].numel() * world_size
+    # Reconstruction protocol: For zero3 we need to zip the partitions together at boundary of each
+    # param, re-consolidating each param, while dealing with padding if any
+
+    # merge list of dicts, preserving order
+    param_shapes = {k: v for d in param_shapes for k, v in d.items()}
+
+    if debug:
+        for i in range(world_size):
+            print(f"{FP32_FLAT_GROUPS}[{i}].shape={fp32_flat_groups[i].shape}")
+
+        wanted_params = len(param_shapes)
+        wanted_numel = sum(shape.numel() for shape in param_shapes.values())
+        # not asserting if there is a mismatch due to possible padding
+        avail_numel = fp32_flat_groups[0].numel() * world_size
+        print(f"Trainable params: Have {avail_numel} numels to process.")
+        print(f"Trainable params: Need {wanted_numel} numels in {wanted_params} params.")
+
+    # params
+    # XXX: for huge models that can't fit into the host's RAM we will have to recode this to support
+    # out-of-core computing solution
+    offset = 0
+    total_numel = 0
+    total_params = 0
+    for name, shape in param_shapes.items():
+
+        unpartitioned_numel = shape.numel()
+        total_numel += unpartitioned_numel
+        total_params += 1
+
+        partitioned_numel, partitioned_padding_numel = zero3_partitioned_param_info(unpartitioned_numel, world_size)
+
+        if debug:
+            print(
+                f"Trainable params: {total_params} {name} full shape: {shape} partition0 numel={partitioned_numel} partitioned_padding_numel={partitioned_padding_numel}"
+            )
+
+        # XXX: memory usage doubles here
+        state_dict[name] = torch.cat(
+            tuple(fp32_flat_groups[i].narrow(0, offset, partitioned_numel) for i in range(world_size)),
+            0).narrow(0, 0, unpartitioned_numel).view(shape)
+        offset += partitioned_numel
+
+    offset *= world_size
+
+    # Sanity check
+    if offset != avail_numel:
+        raise ValueError(f"consumed {offset} numels out of {avail_numel} - something is wrong")
+
+    print(f"Reconstructed Trainable fp32 state dict with {total_params} params {total_numel} elements")
+
+
+def _get_fp32_state_dict_from_zero3_checkpoint(world_size, fp32_flat_groups, zero_model_states):
+    state_dict = OrderedDict()
+
+    # buffers
+    buffers = zero_model_states[0].buffers
+    state_dict.update(buffers)
+    if debug:
+        print(f"added {len(buffers)} buffers")
+
+    _zero3_merge_frozen_params(state_dict, world_size, zero_model_states)
+
+    _zero3_merge_trainable_params(state_dict, world_size, fp32_flat_groups, zero_model_states)
+
+    # recover shared parameters
+    for pair in zero_model_states[0].shared_params:
+        if pair[1] in state_dict:
+            state_dict[pair[0]] = state_dict[pair[1]]
+
+    return state_dict
+
+
+def get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag=None):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated state_dict that can be loaded with
+    ``load_state_dict()`` and used for training without DeepSpeed or shared with others, for example
+    via a model hub.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in 'latest' file. e.g., ``global_step14``
+
+    Returns:
+        - pytorch ``state_dict``
+
+    Note: this approach may not work if your application doesn't have sufficient free CPU memory and
+    you may need to use the offline approach using the ``zero_to_fp32.py`` script that is saved with
+    the checkpoint.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import get_fp32_state_dict_from_zero_checkpoint
+        # do the training and checkpoint saving
+        state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir) # already on cpu
+        model = model.cpu() # move to cpu
+        model.load_state_dict(state_dict)
+        # submit to model hub or save the model to share with others
+
+    In this example the ``model`` will no longer be usable in the deepspeed context of the same
+    application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    If you want it all done for you, use ``load_state_dict_from_zero_checkpoint`` instead.
+
+    """
+    if tag is None:
+        latest_path = os.path.join(checkpoint_dir, 'latest')
+        if os.path.isfile(latest_path):
+            with open(latest_path, 'r') as fd:
+                tag = fd.read().strip()
+        else:
+            raise ValueError(f"Unable to find 'latest' file at {latest_path}")
+
+    ds_checkpoint_dir = os.path.join(checkpoint_dir, tag)
+
+    if not os.path.isdir(ds_checkpoint_dir):
+        raise FileNotFoundError(f"Directory '{ds_checkpoint_dir}' doesn't exist")
+
+    return _get_fp32_state_dict_from_zero_checkpoint(ds_checkpoint_dir)
+
+
+def convert_zero_checkpoint_to_fp32_state_dict(checkpoint_dir, output_file, tag=None):
+    """
+    Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict`` file that can be
+    loaded with ``torch.load(file)`` + ``load_state_dict()`` and used for training without DeepSpeed.
+
+    Args:
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``output_file``: path to the pytorch fp32 state_dict output file (e.g. path/pytorch_model.bin)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+    """
+
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+    print(f"Saving fp32 state dict to {output_file}")
+    torch.save(state_dict, output_file)
+
+
+def load_state_dict_from_zero_checkpoint(model, checkpoint_dir, tag=None):
+    """
+    1. Put the provided model to cpu
+    2. Convert ZeRO 2 or 3 checkpoint into a single fp32 consolidated ``state_dict``
+    3. Load it into the provided model
+
+    Args:
+        - ``model``: the model object to update
+        - ``checkpoint_dir``: path to the desired checkpoint folder. (one that contains the tag-folder, like ``global_step14``)
+        - ``tag``: checkpoint tag used as a unique identifier for checkpoint. If not provided will attempt to load tag in the file named ``latest`` in the checkpoint folder, e.g., ``global_step14``
+
+    Returns:
+        - ``model`: modified model
+
+    Make sure you have plenty of CPU memory available before you call this function. If you don't
+    have enough use the ``zero_to_fp32.py`` utility to do the conversion. You will find it
+    conveniently placed for you in the checkpoint folder.
+
+    A typical usage might be ::
+
+        from deepspeed.utils.zero_to_fp32 import load_state_dict_from_zero_checkpoint
+        model = load_state_dict_from_zero_checkpoint(trainer.model, checkpoint_dir)
+        # submit to model hub or save the model to share with others
+
+    Note, that once this was run, the ``model`` will no longer be usable in the deepspeed context
+    of the same application. i.e. you will need to re-initialize the deepspeed engine, since
+    ``model.load_state_dict(state_dict)`` will remove all the deepspeed magic from it.
+
+    """
+    logger.info(f"Extracting fp32 weights")
+    state_dict = get_fp32_state_dict_from_zero_checkpoint(checkpoint_dir, tag)
+
+    logger.info(f"Overwriting model with fp32 weights")
+    model = model.cpu()
+    model.load_state_dict(state_dict, strict=False)
+
+    return model
+
+
+if __name__ == "__main__":
+
+    parser = argparse.ArgumentParser()
+    parser.add_argument("checkpoint_dir",
+                        type=str,
+                        help="path to the desired checkpoint folder, e.g., path/checkpoint-12")
+    parser.add_argument(
+        "output_file",
+        type=str,
+        help="path to the pytorch fp32 state_dict output file (e.g. path/checkpoint-12/pytorch_model.bin)")
+    parser.add_argument("-t",
+                        "--tag",
+                        type=str,
+                        default=None,
+                        help="checkpoint tag used as a unique identifier for checkpoint. e.g., global_step1")
+    parser.add_argument("-d", "--debug", action='store_true', help="enable debug")
+    args = parser.parse_args()
+
+    debug = args.debug
+
+    convert_zero_checkpoint_to_fp32_state_dict(args.checkpoint_dir, args.output_file, tag=args.tag)

config.json

@@ -0,0 +1,54 @@
+{
+  "_name_or_path": "allenai/OLMo-7B",
+  "activation_type": "swiglu",
+  "alibi": false,
+  "alibi_bias_max": 8.0,
+  "architectures": [
+    "OLMoForCausalLM"
+  ],
+  "attention_dropout": 0.0,
+  "attention_layer_norm": false,
+  "attention_layer_norm_with_affine": false,
+  "auto_map": {
+    "AutoConfig": "allenai/OLMo-7B--configuration_olmo.OLMoConfig",
+    "AutoModelForCausalLM": "allenai/OLMo-7B--modeling_olmo.OLMoForCausalLM",
+    "AutoTokenizer": [
+      "allenai/OLMo-7B--tokenization_olmo_fast.OLMoTokenizerFast",
+      "allenai/OLMo-7B--tokenization_olmo_fast.OLMoTokenizerFast"
+    ]
+  },
+  "bias_for_layer_norm": false,
+  "block_group_size": 1,
+  "block_type": "sequential",
+  "clip_qkv": null,
+  "d_model": 4096,
+  "embedding_dropout": 0.0,
+  "embedding_size": 50304,
+  "eos_token_id": 50279,
+  "flash_attention": true,
+  "include_bias": false,
+  "init_cutoff_factor": null,
+  "init_device": "meta",
+  "init_fn": "mitchell",
+  "init_std": 0.02,
+  "layer_norm_type": "default",
+  "layer_norm_with_affine": false,
+  "max_sequence_length": 4096,
+  "mlp_hidden_size": 22016,
+  "mlp_ratio": 4,
+  "model_type": "olmo",
+  "multi_query_attention": false,
+  "n_heads": 32,
+  "n_layers": 32,
+  "pad_token_id": 1,
+  "precision": "amp_bf16",
+  "residual_dropout": 0.0,
+  "rope": true,
+  "rope_full_precision": true,
+  "scale_logits": false,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.38.0",
+  "use_cache": false,
+  "vocab_size": 50280,
+  "weight_tying": false
+}

generation_config.json

@@ -0,0 +1,7 @@
+{
+  "_from_model_config": true,
+  "do_sample": true,
+  "eos_token_id": 50279,
+  "pad_token_id": 1,
+  "transformers_version": "4.38.0"
+}

pytorch_model.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c87135fefaebb593d88b82da343be87428e8b03b8a4fea367bb8c0925999689b
+size 13776210149

special_tokens_map.json

@@ -0,0 +1,16 @@
+{
+  "eos_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|padding|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer_config.json

@@ -0,0 +1,242 @@
+{
+  "added_tokens_decoder": {
+    "0": {
+      "content": "|||IP_ADDRESS|||",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "1": {
+      "content": "<|padding|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "50254": {
+      "content": "                        ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50255": {
+      "content": "                       ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50256": {
+      "content": "                      ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50257": {
+      "content": "                     ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50258": {
+      "content": "                    ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50259": {
+      "content": "                   ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50260": {
+      "content": "                  ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50261": {
+      "content": "                 ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50262": {
+      "content": "                ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50263": {
+      "content": "               ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50264": {
+      "content": "              ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50265": {
+      "content": "             ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50266": {
+      "content": "            ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50267": {
+      "content": "           ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50268": {
+      "content": "          ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50269": {
+      "content": "         ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50270": {
+      "content": "        ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50271": {
+      "content": "       ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50272": {
+      "content": "      ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50273": {
+      "content": "     ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50274": {
+      "content": "    ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50275": {
+      "content": "   ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50276": {
+      "content": "  ",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50277": {
+      "content": "|||EMAIL_ADDRESS|||",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50278": {
+      "content": "|||PHONE_NUMBER|||",
+      "lstrip": false,
+      "normalized": true,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "50279": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    }
+  },
+  "auto_map": {
+    "AutoConfig": "allenai/OLMo-7B--configuration_olmo.OLMoConfig",
+    "AutoTokenizer": [
+      "allenai/OLMo-7B--tokenization_olmo_fast.OLMoTokenizerFast",
+      "allenai/OLMo-7B--tokenization_olmo_fast.OLMoTokenizerFast"
+    ]
+  },
+  "clean_up_tokenization_spaces": true,
+  "eos_token": "<|endoftext|>",
+  "max_length": null,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<|padding|>",
+  "tokenizer_class": "OLMoTokenizer",
+  "truncation": "right"
+}