diff --git a/.gitattributes b/.gitattributes
index a6344aac8c09253b3b630fb776ae94478aa0275b..8c5681d4d8e3536579d160928c07dfc7cff10ce8 100644
--- a/.gitattributes
+++ b/.gitattributes
@@ -33,3 +33,132 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/.metadata filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__0_0.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__0_1.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__0_10.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__0_11.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__0_12.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__0_13.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__0_14.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__0_15.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__0_2.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__0_3.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__0_4.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__0_5.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__0_6.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__0_7.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__0_8.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__0_9.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__1_0.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__1_1.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__1_10.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__1_11.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__1_12.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__1_13.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__1_14.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__1_15.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__1_2.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__1_3.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__1_4.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__1_5.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__1_6.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__1_7.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__1_8.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__1_9.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__2_0.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__2_1.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__2_10.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__2_11.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__2_12.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__2_13.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__2_14.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__2_15.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__2_2.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__2_3.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__2_4.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__2_5.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__2_6.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__2_7.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__2_8.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__2_9.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__3_0.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__3_1.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__3_10.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__3_11.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__3_12.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__3_13.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__3_14.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__3_15.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__3_2.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__3_3.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__3_4.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__3_5.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__3_6.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__3_7.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__3_8.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__3_9.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__4_0.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__4_1.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__4_10.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__4_11.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__4_12.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__4_13.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__4_14.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__4_15.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__4_2.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__4_3.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__4_4.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__4_5.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__4_6.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__4_7.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__4_8.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__4_9.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__5_0.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__5_1.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__5_10.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__5_11.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__5_12.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__5_13.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__5_14.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__5_15.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__5_2.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__5_3.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__5_4.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__5_5.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__5_6.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__5_7.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__5_8.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__5_9.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__6_0.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__6_1.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__6_10.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__6_11.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__6_12.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__6_13.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__6_14.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__6_15.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__6_2.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__6_3.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__6_4.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__6_5.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__6_6.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__6_7.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__6_8.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__6_9.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__7_0.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__7_1.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__7_10.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__7_11.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__7_12.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__7_13.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__7_14.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__7_15.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__7_2.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__7_3.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__7_4.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__7_5.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__7_6.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__7_7.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__7_8.distcp filter=lfs diff=lfs merge=lfs -text
+olmo_core/model_and_optim/__7_9.distcp filter=lfs diff=lfs merge=lfs -text
diff --git a/olmo_core/.metadata.json b/olmo_core/.metadata.json
new file mode 100644
index 0000000000000000000000000000000000000000..a410f3651cf31f208c4ff898cd4f43126c864fff
--- /dev/null
+++ b/olmo_core/.metadata.json
@@ -0,0 +1 @@
+{"version": "2.3.0"}
\ No newline at end of file
diff --git a/olmo_core/config.json b/olmo_core/config.json
new file mode 100644
index 0000000000000000000000000000000000000000..f79d4f46a054097c29c33e65b4d002e6ff16d510
--- /dev/null
+++ b/olmo_core/config.json
@@ -0,0 +1,349 @@
+{
+    "model": {
+        "d_model": 2048,
+        "vocab_size": 640,
+        "n_layers": 16,
+        "block": {
+            "attention": {
+                "name": "default",
+                "n_heads": 16,
+                "bias": false,
+                "rope": {
+                    "name": "default",
+                    "theta": 500000,
+                    "full_precision": true,
+                    "_CLASS_": "olmo_core.nn.rope.RoPEConfig"
+                },
+                "qk_norm": {
+                    "name": "rms",
+                    "eps": 1e-06,
+                    "bias": false,
+                    "dtype": "float32",
+                    "_CLASS_": "olmo_core.nn.layer_norm.LayerNormConfig"
+                },
+                "use_flash": true,
+                "dtype": "float32",
+                "_CLASS_": "olmo_core.nn.attention.AttentionConfig"
+            },
+            "layer_norm": {
+                "name": "rms",
+                "eps": 1e-06,
+                "bias": false,
+                "dtype": "float32",
+                "_CLASS_": "olmo_core.nn.layer_norm.LayerNormConfig"
+            },
+            "feed_forward": {
+                "hidden_size": 8192,
+                "name": "default",
+                "bias": false,
+                "dtype": "float32",
+                "act_name": "silu",
+                "_CLASS_": "olmo_core.nn.feed_forward.FeedForwardConfig"
+            },
+            "name": "reordered_norm",
+            "_CLASS_": "olmo_core.nn.transformer.config.TransformerBlockConfig"
+        },
+        "lm_head": {
+            "name": "default",
+            "layer_norm": {
+                "name": "rms",
+                "eps": 1e-06,
+                "bias": false,
+                "dtype": "float32",
+                "_CLASS_": "olmo_core.nn.layer_norm.LayerNormConfig"
+            },
+            "bias": false,
+            "dtype": "float32",
+            "loss_implementation": "default",
+            "_CLASS_": "olmo_core.nn.lm_head.LMHeadConfig"
+        },
+        "name": "bolmo_distill",
+        "dtype": "float32",
+        "init_method": "normal",
+        "init_seed": 0,
+        "init_std": 0.02,
+        "freeze_params": [
+            "boundary_predictor.*",
+            "teacher_embeddings.*"
+        ],
+        "local_encoder": {
+            "sliding_window_size": 0,
+            "d_model": 2048,
+            "n_layers": 1,
+            "block_config": {
+                "attention": {
+                    "name": "default",
+                    "n_heads": 16,
+                    "dtype": "float32",
+                    "_CLASS_": "olmo_core.nn.attention.AttentionConfig"
+                },
+                "layer_norm": {
+                    "name": "rms",
+                    "eps": 1e-06,
+                    "bias": false,
+                    "dtype": "float32",
+                    "_CLASS_": "olmo_core.nn.layer_norm.LayerNormConfig"
+                },
+                "feed_forward": {
+                    "hidden_size": 2816,
+                    "name": "default",
+                    "bias": false,
+                    "dtype": "float32",
+                    "act_name": "silu",
+                    "_CLASS_": "olmo_core.nn.feed_forward.FeedForwardConfig"
+                },
+                "xlstm": {
+                    "num_heads": 16,
+                    "dtype": "float32",
+                    "_CLASS_": "olmo_core.nn.xlstm.XLSTMConfig"
+                },
+                "name": "xlstm",
+                "_CLASS_": "olmo_core.nn.transformer.config.TransformerBlockConfig"
+            },
+            "cross_attn_n_heads": 0,
+            "cross_attn_do_project": true,
+            "cross_attn_init_pooling": "amax",
+            "pooling": "hnet",
+            "add_hash_embeddings": false,
+            "add_expanded_embeddings": true,
+            "hash_byte_group_size": [
+                3,
+                4,
+                5,
+                6,
+                7,
+                8
+            ],
+            "hash_byte_group_vocab": [
+                1536,
+                3072,
+                6144,
+                12288,
+                24576,
+                49152
+            ],
+            "hash_byte_group_nb_functions": 1,
+            "add_norm_after_last_block": true,
+            "add_norm_after_pool": false,
+            "add_out_projection": true,
+            "boundary_predictor": "hnet",
+            "boundary_predictor_lookahead": 1,
+            "represent_bytes_with_embeddings": false,
+            "represent_bytes_with_last_mixed_out": false,
+            "blt_compat": false,
+            "dtype": "float32",
+            "_CLASS_": "olmo_core.nn.bolmo.config.LocalEncoderConfig"
+        },
+        "local_decoder": {
+            "sliding_window_size": 0,
+            "d_model": 2048,
+            "n_layers": 4,
+            "cross_attn_n_heads": 0,
+            "block_config": {
+                "attention": {
+                    "name": "default",
+                    "n_heads": 16,
+                    "dtype": "float32",
+                    "_CLASS_": "olmo_core.nn.attention.AttentionConfig"
+                },
+                "layer_norm": {
+                    "name": "rms",
+                    "eps": 1e-06,
+                    "bias": false,
+                    "dtype": "float32",
+                    "_CLASS_": "olmo_core.nn.layer_norm.LayerNormConfig"
+                },
+                "feed_forward": {
+                    "hidden_size": 2816,
+                    "name": "default",
+                    "bias": false,
+                    "dtype": "float32",
+                    "act_name": "silu",
+                    "_CLASS_": "olmo_core.nn.feed_forward.FeedForwardConfig"
+                },
+                "xlstm": {
+                    "num_heads": 16,
+                    "dtype": "float32",
+                    "_CLASS_": "olmo_core.nn.xlstm.XLSTMConfig"
+                },
+                "name": "xlstm",
+                "_CLASS_": "olmo_core.nn.transformer.config.TransformerBlockConfig"
+            },
+            "depooling": "hnet",
+            "add_norm_before_first_block": true,
+            "add_norm_onto_residual": false,
+            "add_in_projection": true,
+            "add_projected_patch_residuals": false,
+            "hnet_smooth": false,
+            "hnet_smooth_ste": false,
+            "hnet_modulate": false,
+            "blt_compat": false,
+            "fuse_boundaries": true,
+            "no_boundaries": false,
+            "dtype": "float32",
+            "_CLASS_": "olmo_core.nn.bolmo.config.LocalDecoderConfig"
+        },
+        "share_blocks_between_teacher_and_student": false,
+        "_CLASS_": "olmo_core.nn.transformer.config.TransformerConfig"
+    },
+    "dataset": {
+        "tokenizer": {
+            "vocab_size": 520,
+            "eos_token_id": 1,
+            "pad_token_id": 0,
+            "bos_token_id": 1,
+            "special_tokens": [
+                "<pad>",
+                "<bos>",
+                "<eos>",
+                "<bpe_token_end>"
+            ],
+            "special_tokens_first": true,
+            "original_identifier": "allenai/dolma2-tokenizer",
+            "bpe_token_end_id": 3,
+            "_CLASS_": "olmo_core.data.tokenizer.ByteTokenizerConfig"
+        },
+        "paths": [],
+        "expand_glob": false,
+        "include_instance_metadata": true,
+        "work_dir": "",
+        "ignore_fingerprint_mismatch": false,
+        "sequence_length": 4096,
+        "generate_doc_lengths": false,
+        "byte_sequence_length": 24576,
+        "_CLASS_": "olmo_core.data.numpy_dataset.NumpyByteFSLDatasetConfig"
+    },
+    "data_loader": {
+        "global_batch_size": 1572864,
+        "seed": 1234,
+        "num_workers": 24,
+        "ignore_fingerprint_mismatch": false,
+        "_CLASS_": "olmo_core.data.data_loader.NumpyDataLoaderConfig"
+    },
+    "train_module": {
+        "rank_microbatch_size": 98304,
+        "max_sequence_length": 24576,
+        "optim": {
+            "group_overrides": [
+                {
+                    "params": [
+                        "local_encoder.embedding.weight",
+                        "local_encoder.expanded_embeddings.weight"
+                    ],
+                    "opts": {
+                        "weight_decay": 0.0
+                    },
+                    "_CLASS_": "olmo_core.optim.config.OptimGroupOverride"
+                },
+                {
+                    "params": [
+                        "blocks.*"
+                    ],
+                    "opts": {
+                        "lr": 2.6e-05
+                    },
+                    "_CLASS_": "olmo_core.optim.config.OptimGroupOverride"
+                }
+            ],
+            "compile": false,
+            "fixed_fields": [
+                "initial_lr"
+            ],
+            "lr": 5.2e-05,
+            "betas": [
+                0.9,
+                0.95
+            ],
+            "eps": 1e-08,
+            "weight_decay": 0.1,
+            "_CLASS_": "olmo_core.optim.adamw.AdamWConfig"
+        },
+        "max_grad_norm": 0.5,
+        "scheduler": {
+            "lr_field": "lr",
+            "initial_lr_field": "initial_lr",
+            "units": "steps",
+            "alpha_f": 0.0,
+            "warmup_fraction": 0.1,
+            "warmup_min_lr": 0.0,
+            "_CLASS_": "olmo_core.optim.scheduler.LinearWithWarmup"
+        },
+        "compile_model": true,
+        "float8_config": {
+            "enabled": false,
+            "_CLASS_": "olmo_core.float8.Float8Config"
+        },
+        "dp_config": {
+            "name": "fsdp",
+            "param_dtype": "bfloat16",
+            "reduce_dtype": "float32",
+            "wrapping_strategy": "full",
+            "prefetch_factor": 0,
+            "_CLASS_": "olmo_core.train.train_module.transformer.config.TransformerDataParallelConfig"
+        },
+        "bolmo_config": {
+            "tokenizer": {
+                "vocab_size": 520,
+                "eos_token_id": 1,
+                "pad_token_id": 0,
+                "bos_token_id": 1,
+                "special_tokens": [
+                    "<pad>",
+                    "<bos>",
+                    "<eos>",
+                    "<bpe_token_end>"
+                ],
+                "special_tokens_first": true,
+                "original_identifier": "allenai/dolma2-tokenizer",
+                "bpe_token_end_id": 3,
+                "_CLASS_": "olmo_core.data.tokenizer.ByteTokenizerConfig"
+            },
+            "losses": [
+                "ce",
+                "boundary"
+            ],
+            "loss_weights": [
+                1.0,
+                4.0
+            ],
+            "binarization_temp": 1.0,
+            "temperature": 1.0,
+            "div_fn": "tvd_temp_limit",
+            "boundary_mode": "end",
+            "merge_boundary_loss": false,
+            "use_output_boundary_jsd": false,
+            "eval_add_boundary_logp": false,
+            "do_alm_debiasing": false,
+            "rep_compare_fn": "l2",
+            "start_ratio": 4.3,
+            "target_ratio": 4.3,
+            "gradual_boundary_compression_steps": 150000,
+            "encoder_loss_lookahead": 0,
+            "encoder_loss_no_lookahead_weight": 1.0,
+            "encoder_loss_lookahead_weights": [],
+            "patching": "dolma2",
+            "epsilon": 1e-06,
+            "skip_blocks": false,
+            "skip_teacher_blocks": false,
+            "skip_teacher": true,
+            "compute_teacher_ce": false,
+            "use_student_patch_reps_for_teacher": false,
+            "use_oracle_patch_reps": false,
+            "teacher_blocks_no_grad": true,
+            "student_blocks_no_grad": false,
+            "decoder_backprop_through_encoder": true,
+            "decoder_backprop_through_boundary_predictor": true,
+            "boundary_predictor_backprop_through_encoder": true,
+            "teacher_force_boundaries": false,
+            "boundary_threshold": "sample:0",
+            "xlstm_igate_bias_init": -10.0,
+            "skip_boundary_before_eos": true,
+            "_CLASS_": "olmo_core.nn.bolmo.config.BolmoConfig"
+        },
+        "label_ignore_index": -100,
+        "_CLASS_": "olmo_core.train.train_module.transformer.config.TransformerTrainModuleConfig"
+    },
+    "trainer": {},
+    "init_seed": 12536,
+    "_CLASS_": "__main__.ExperimentConfig"
+}
\ No newline at end of file
diff --git a/olmo_core/model_and_optim/.metadata b/olmo_core/model_and_optim/.metadata
new file mode 100644
index 0000000000000000000000000000000000000000..f67cc8ed9cde8c77189b6c40b2575ae409d9c474
--- /dev/null
+++ b/olmo_core/model_and_optim/.metadata
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ea8ac9dd55708f4a36911565f33e7e223a9423ef21c79b868869c14bbd0b6cfa
+size 2464795
diff --git a/olmo_core/model_and_optim/__0_0.distcp b/olmo_core/model_and_optim/__0_0.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..70cf2d85aaaf21a056131158b3b619cad78f70e3
--- /dev/null
+++ b/olmo_core/model_and_optim/__0_0.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9241563abb2e63f5e76e9047b32d1b46adf82bc18bba2fc33dcff6c1caf4e856
+size 138224729
diff --git a/olmo_core/model_and_optim/__0_1.distcp b/olmo_core/model_and_optim/__0_1.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..321e1e370ec1dc388c2f3cc78a01b4591f470db9
--- /dev/null
+++ b/olmo_core/model_and_optim/__0_1.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6ced12b46582166b9e74e636518a332aff6fa25ca3295baa81a9fb61d7844ac
+size 138224729
diff --git a/olmo_core/model_and_optim/__0_10.distcp b/olmo_core/model_and_optim/__0_10.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..fd63e8c885be221f55117c27a0337632da5a3dff
--- /dev/null
+++ b/olmo_core/model_and_optim/__0_10.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7afb9d5ccf88ad2e9c0f3a4db73aab41f04065731bef1cb9e3e632a2d32a3bd
+size 138054699
diff --git a/olmo_core/model_and_optim/__0_11.distcp b/olmo_core/model_and_optim/__0_11.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..f596da5993b258c85c200819218340090c1b5cce
--- /dev/null
+++ b/olmo_core/model_and_optim/__0_11.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0285c42fd4de086b601dacbcc1b6042c30156a2ad0569c4c49e3864f99fdbb75
+size 138053122
diff --git a/olmo_core/model_and_optim/__0_12.distcp b/olmo_core/model_and_optim/__0_12.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..b9cede312cfe86ec3553ba0eaea9e7b22df4d947
--- /dev/null
+++ b/olmo_core/model_and_optim/__0_12.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d727be4cca2724990cfb0f8e45ee9d71538a48b93a812f26246a3112391d55a0
+size 137924481
diff --git a/olmo_core/model_and_optim/__0_13.distcp b/olmo_core/model_and_optim/__0_13.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..9097b705f42b9c610ca4363163113c59d0717573
--- /dev/null
+++ b/olmo_core/model_and_optim/__0_13.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c10f064d3c4a5a8f3b2020a8e630bc0270bb6dc50e7de8ea32a6c23e0f917ba5
+size 137924481
diff --git a/olmo_core/model_and_optim/__0_14.distcp b/olmo_core/model_and_optim/__0_14.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..d7625e6af4a21d0c1fc0c529f0573c6c01f0f178
--- /dev/null
+++ b/olmo_core/model_and_optim/__0_14.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f66920be4bb54652f8ebb8db3b01b448ad6ff467e2caf300dd74df04e780e50f
+size 137926058
diff --git a/olmo_core/model_and_optim/__0_15.distcp b/olmo_core/model_and_optim/__0_15.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..89b037c02c282f8d4b4cfd5a010760eefea8f34b
--- /dev/null
+++ b/olmo_core/model_and_optim/__0_15.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:89eececcc6343abcd58709df669e959793180a439bedd5a9269028997c3b95f4
+size 137926058
diff --git a/olmo_core/model_and_optim/__0_2.distcp b/olmo_core/model_and_optim/__0_2.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..7669bcbe85d57817200226bd821e394afcbdd8dc
--- /dev/null
+++ b/olmo_core/model_and_optim/__0_2.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:452c356d9c7dfca4d617d3c41e2b758c516515e4319149770c9d4cee22751a98
+size 138224729
diff --git a/olmo_core/model_and_optim/__0_3.distcp b/olmo_core/model_and_optim/__0_3.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..d4b6d19ee6ea9cf84a08b475157aaf74e3d042d2
--- /dev/null
+++ b/olmo_core/model_and_optim/__0_3.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:93056e9e58af6bf3efce9901b41424a7bea1f3937c4fa69567d7ff904ec9cc3b
+size 138448453
diff --git a/olmo_core/model_and_optim/__0_4.distcp b/olmo_core/model_and_optim/__0_4.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..d3abf269d9d568d232dd6408d523fa7be90939c4
--- /dev/null
+++ b/olmo_core/model_and_optim/__0_4.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:165e64ab61aac760c9ac06bc259d989f7ba6caf9dc80a320040e2094ac4031c7
+size 138319274
diff --git a/olmo_core/model_and_optim/__0_5.distcp b/olmo_core/model_and_optim/__0_5.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..6a75866344732f42de2bd9cac27351453d256c35
--- /dev/null
+++ b/olmo_core/model_and_optim/__0_5.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:06bb7b56204a7491b3471bf4ce34eebe72a483fea29dd2ca6d950a1b69c5a4b2
+size 138319274
diff --git a/olmo_core/model_and_optim/__0_6.distcp b/olmo_core/model_and_optim/__0_6.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..3eefbd433df961e2c26e50aaa6143886c9cccd6d
--- /dev/null
+++ b/olmo_core/model_and_optim/__0_6.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8febe1110dd2aa1c63e8f4112934e0432e7cb120ad379705b6bd6f69c900fff4
+size 138319274
diff --git a/olmo_core/model_and_optim/__0_7.distcp b/olmo_core/model_and_optim/__0_7.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..ac7ce8af3a428261211ad78e93ad2a4d9edbca4f
--- /dev/null
+++ b/olmo_core/model_and_optim/__0_7.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:29f5377b1855852505822287811f35dca870bd9a89711c6dd91e053917cfc798
+size 137902754
diff --git a/olmo_core/model_and_optim/__0_8.distcp b/olmo_core/model_and_optim/__0_8.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..a7dbe49a1c58012f5066cb841245551048adc894
--- /dev/null
+++ b/olmo_core/model_and_optim/__0_8.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e6943898c5bb21c2b2249dad0168defbfbdcb63a7ad0973d751b468ec07d3f40
+size 137902754
diff --git a/olmo_core/model_and_optim/__0_9.distcp b/olmo_core/model_and_optim/__0_9.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..2e0ca66ac8f686b94637d02772e9026de14d6097
--- /dev/null
+++ b/olmo_core/model_and_optim/__0_9.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d46b7bc5207027ea7fd5d98242a338a2bc0c70dd3350d2e39851ca7e6688dfc1
+size 138054699
diff --git a/olmo_core/model_and_optim/__1_0.distcp b/olmo_core/model_and_optim/__1_0.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..3ded48649fc62bcf15a8dfd56474ff1e08e64b65
--- /dev/null
+++ b/olmo_core/model_and_optim/__1_0.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d0e30b56e0f4cdea3fcad9e636a0131caf679e2a5d66ad69fca0d60608001ce3
+size 137975051
diff --git a/olmo_core/model_and_optim/__1_1.distcp b/olmo_core/model_and_optim/__1_1.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..668a8baddfc21f6a9c0b2b80d3adcb412ab55491
--- /dev/null
+++ b/olmo_core/model_and_optim/__1_1.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f19ba4c47b4093ee13f16632f9be4f04af4f068b7462db621139bbf1c2ba2678
+size 137975051
diff --git a/olmo_core/model_and_optim/__1_10.distcp b/olmo_core/model_and_optim/__1_10.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..b8aa98b7f322033150c85d1eb84078193bf337a7
--- /dev/null
+++ b/olmo_core/model_and_optim/__1_10.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fd8a8524fca46aea6135faec75282ecae3ecc451b0f693614192f0b8a19a59c1
+size 137667506
diff --git a/olmo_core/model_and_optim/__1_11.distcp b/olmo_core/model_and_optim/__1_11.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..1a992265bfea05b0c9e1038ae94ad7a411bce074
--- /dev/null
+++ b/olmo_core/model_and_optim/__1_11.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:cf27a46ff7598b81b3bd5f8663147542d7a16d6d31a44ed20968d06ff2b334c2
+size 137667506
diff --git a/olmo_core/model_and_optim/__1_12.distcp b/olmo_core/model_and_optim/__1_12.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..7a43ab9c2f927dcd02650e33725f54e47c9a451e
--- /dev/null
+++ b/olmo_core/model_and_optim/__1_12.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c7de5fbb6ddd3855373b1311eb09827f7bce000b9cec11255f10ee17fcbf05eb
+size 137676064
diff --git a/olmo_core/model_and_optim/__1_13.distcp b/olmo_core/model_and_optim/__1_13.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..b39b1581fb4ce933a7dcd79966413c8500367ef6
--- /dev/null
+++ b/olmo_core/model_and_optim/__1_13.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c6ec75d600a68102f796d9d8fce2fdb0b9898dff43595cad570b60efcdb269b0
+size 137676064
diff --git a/olmo_core/model_and_optim/__1_14.distcp b/olmo_core/model_and_optim/__1_14.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..5b81f5f0ac2660ead94acc88cc756feb67465713
--- /dev/null
+++ b/olmo_core/model_and_optim/__1_14.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9b66435c1ed78293f1af1f31bedbd97033cc797597391ab045bc584d58797f58
+size 137677641
diff --git a/olmo_core/model_and_optim/__1_15.distcp b/olmo_core/model_and_optim/__1_15.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..16449d39b20d0dce6549d3a8a8dd120a7db43684
--- /dev/null
+++ b/olmo_core/model_and_optim/__1_15.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ff36c0a40ac9b5137b10958759d3c3eb7fe1f64d0d9699ef27f44d1a7104ee43
+size 137677641
diff --git a/olmo_core/model_and_optim/__1_2.distcp b/olmo_core/model_and_optim/__1_2.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..fc4771c6d12b5a397e1b56166dacf3fe570457f2
--- /dev/null
+++ b/olmo_core/model_and_optim/__1_2.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:efb835fa07e895daeef6b90d4528fe24de85f1d3950db6b123342594bcd41387
+size 137975051
diff --git a/olmo_core/model_and_optim/__1_3.distcp b/olmo_core/model_and_optim/__1_3.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..5622958088fe82d92e61e70cb583a9d75246fba0
--- /dev/null
+++ b/olmo_core/model_and_optim/__1_3.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:67679cdff1de9eb8b5574a2941ada5f5d98e7e3f7d8d8fac8cf6e934879db0aa
+size 138198775
diff --git a/olmo_core/model_and_optim/__1_4.distcp b/olmo_core/model_and_optim/__1_4.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..ab3dc126d613e1cfe3a33234ded893c7c0b60193
--- /dev/null
+++ b/olmo_core/model_and_optim/__1_4.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:27723cefd17c891e7d891610acffeaf7ac2e8d31f109118b92bfee9c55016ba0
+size 138070857
diff --git a/olmo_core/model_and_optim/__1_5.distcp b/olmo_core/model_and_optim/__1_5.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..6ceb879c6acc573dad6cc627e0718fa4d0663db4
--- /dev/null
+++ b/olmo_core/model_and_optim/__1_5.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ad447e78e2efadf7eab55bac518eb88714645b477a93f222048b5a856f7b15c0
+size 138070857
diff --git a/olmo_core/model_and_optim/__1_6.distcp b/olmo_core/model_and_optim/__1_6.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..c658a8041a25fa0f38522d6566718adb9af0660b
--- /dev/null
+++ b/olmo_core/model_and_optim/__1_6.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8efabe23bc8209221a31ec2c2c0dd0ac4a24444cfb53eea1423d402406891cb7
+size 138070857
diff --git a/olmo_core/model_and_optim/__1_7.distcp b/olmo_core/model_and_optim/__1_7.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..db03b00c7ea70a4e920996003823d0b0b8f534e3
--- /dev/null
+++ b/olmo_core/model_and_optim/__1_7.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:340d3f0e71d0e35e23a7a34c035339edfecb1268b9fdbcb6370e18be2754adc0
+size 137654337
diff --git a/olmo_core/model_and_optim/__1_8.distcp b/olmo_core/model_and_optim/__1_8.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..d5c1689940c3ec31e50c05019a3c4bba01998245
--- /dev/null
+++ b/olmo_core/model_and_optim/__1_8.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d7b4d3f88074fd110a32ca232d5264a3848f8caa11f5334ff71c74df8e999d00
+size 137654337
diff --git a/olmo_core/model_and_optim/__1_9.distcp b/olmo_core/model_and_optim/__1_9.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..594e32d3ebbccee041fa6f1f2551e2131f782a26
--- /dev/null
+++ b/olmo_core/model_and_optim/__1_9.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:118cdbe40e52906b81456b66df6d4a5743a9183f6dffc9afbba01a596a547faa
+size 137667506
diff --git a/olmo_core/model_and_optim/__2_0.distcp b/olmo_core/model_and_optim/__2_0.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..4c4499e08036a52c42f15f73cdb0cedaf893477c
--- /dev/null
+++ b/olmo_core/model_and_optim/__2_0.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9f6a41c5048cde6511c0b50bba5684c39f6922fafb5b11931e01a2a4160d8604
+size 137975051
diff --git a/olmo_core/model_and_optim/__2_1.distcp b/olmo_core/model_and_optim/__2_1.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..6646bb10a1cb450fd6201da8b8d103d6ed390c49
--- /dev/null
+++ b/olmo_core/model_and_optim/__2_1.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:328dd39d60e3663225c6c6fdb38e9aa600426dca217c0d962208e7af954d49f1
+size 137975051
diff --git a/olmo_core/model_and_optim/__2_10.distcp b/olmo_core/model_and_optim/__2_10.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..1dca289d103e274be4b5250cc5284d8d968531fb
--- /dev/null
+++ b/olmo_core/model_and_optim/__2_10.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0578b542929035283b915bb5ec4883c42b9dd801f7988b3bef83033816d9ed35
+size 137667506
diff --git a/olmo_core/model_and_optim/__2_11.distcp b/olmo_core/model_and_optim/__2_11.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..55304517918ec597733d8819162ca08b4d29ee85
--- /dev/null
+++ b/olmo_core/model_and_optim/__2_11.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da49896618a8f8303b681ec064c6355088b58fafd33b4c62b6cc740f1038851f
+size 137667506
diff --git a/olmo_core/model_and_optim/__2_12.distcp b/olmo_core/model_and_optim/__2_12.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..12c78d790a9d2ff342498f8784432faf66ac51e6
--- /dev/null
+++ b/olmo_core/model_and_optim/__2_12.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c79ff34c25d25fe399d9a289fcf6d1031142a88d62c50dc50b150aa02aa4956
+size 137676064
diff --git a/olmo_core/model_and_optim/__2_13.distcp b/olmo_core/model_and_optim/__2_13.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..e8887fb9d1e847777fbba1299fd36012bc8f7ee8
--- /dev/null
+++ b/olmo_core/model_and_optim/__2_13.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fb401a50823a9ed8038e1968d9a3e222d05509c33079c2b8b16e282cd67f5b41
+size 137676064
diff --git a/olmo_core/model_and_optim/__2_14.distcp b/olmo_core/model_and_optim/__2_14.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..4dcb99a7ae557f340780bcf810a54d382b713401
--- /dev/null
+++ b/olmo_core/model_and_optim/__2_14.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:eb63c9c083f627e150c041a2d3de9282d6f643dae1e95e1a46691684fe406459
+size 137677641
diff --git a/olmo_core/model_and_optim/__2_15.distcp b/olmo_core/model_and_optim/__2_15.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..42af2b36dde99fc5bf8a7733871e7c84271ecec2
--- /dev/null
+++ b/olmo_core/model_and_optim/__2_15.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1205969c3caa7aaf7073f5afb49061fc7d646f86704cd4fab2caf44d28c00a6
+size 137677641
diff --git a/olmo_core/model_and_optim/__2_2.distcp b/olmo_core/model_and_optim/__2_2.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..dc3566609fe95e6197efe714634afbf83247edfc
--- /dev/null
+++ b/olmo_core/model_and_optim/__2_2.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:51c4181a8c59c612f46ff9ce6520da3c74f841bc20a137ddac9b2d2e5f3f7690
+size 137975051
diff --git a/olmo_core/model_and_optim/__2_3.distcp b/olmo_core/model_and_optim/__2_3.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..3c6d66ccd45100b26edad3370952f1ac3c66a7c5
--- /dev/null
+++ b/olmo_core/model_and_optim/__2_3.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:94dd1a66905a5908f094413497af69d448d1d38159a2f97dfe3e78143f892c14
+size 138198775
diff --git a/olmo_core/model_and_optim/__2_4.distcp b/olmo_core/model_and_optim/__2_4.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..ede01687da069080d9da1b1f3858bcd8a9c51bbc
--- /dev/null
+++ b/olmo_core/model_and_optim/__2_4.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:daf72d44ed90913fce12cfbe2626044c68edf87b14f5ebb3a48eb77591a80447
+size 138070857
diff --git a/olmo_core/model_and_optim/__2_5.distcp b/olmo_core/model_and_optim/__2_5.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..d0e0859e80a0ecee17e011c299ff5b2433a30a1f
--- /dev/null
+++ b/olmo_core/model_and_optim/__2_5.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a6fcb095aa0987410995ae325c87eb5e01bcdf0dfa35eae629e1ef2a11a76b3f
+size 138070857
diff --git a/olmo_core/model_and_optim/__2_6.distcp b/olmo_core/model_and_optim/__2_6.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..3db570cbce559908dd39f9c30eb1579851968aae
--- /dev/null
+++ b/olmo_core/model_and_optim/__2_6.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c8c5f22aabf49f0afc77bd3871ad5939f429d088814ede8bd602ae7f9e52055
+size 138070857
diff --git a/olmo_core/model_and_optim/__2_7.distcp b/olmo_core/model_and_optim/__2_7.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..4897c30486b20b7f183872684aeb0c221000f08f
--- /dev/null
+++ b/olmo_core/model_and_optim/__2_7.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bc9faf4b47e1ea02e5a1b0fa8e8cddea7ac92bda6d230945722d594ece8b82bc
+size 137654337
diff --git a/olmo_core/model_and_optim/__2_8.distcp b/olmo_core/model_and_optim/__2_8.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..2e6ad92357612e9442233fc4c7208bc3f277dd71
--- /dev/null
+++ b/olmo_core/model_and_optim/__2_8.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1895c32163097f7de47f8aafeca16e2da02d20f4548facf9be8c37429d85fec4
+size 137654337
diff --git a/olmo_core/model_and_optim/__2_9.distcp b/olmo_core/model_and_optim/__2_9.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..96cf67307efc7d2310b86845e54e6c566120d2a9
--- /dev/null
+++ b/olmo_core/model_and_optim/__2_9.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:36a6ef73ac8d09189b6a5624b25da0b47a17301f482c4663a74af1a2a0cd34b8
+size 137667506
diff --git a/olmo_core/model_and_optim/__3_0.distcp b/olmo_core/model_and_optim/__3_0.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..d539aa03c0e0f06741f50a63c9fc86f7cea8a9b9
--- /dev/null
+++ b/olmo_core/model_and_optim/__3_0.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:240c26e68c3202fbae697a1a5a2afadd21c797ee97d4dbe66b48935e3ec8f716
+size 137975051
diff --git a/olmo_core/model_and_optim/__3_1.distcp b/olmo_core/model_and_optim/__3_1.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..745cc416c6f0f97b8b3b7e25b8bf7ebd7be9cfea
--- /dev/null
+++ b/olmo_core/model_and_optim/__3_1.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:95ade36b14369c3975a08c0fa9ff359631d7be423705f19a7fca76f49728f941
+size 137975051
diff --git a/olmo_core/model_and_optim/__3_10.distcp b/olmo_core/model_and_optim/__3_10.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..b8720be5a769f7d3437dd226b2aff4b4dff3b45e
--- /dev/null
+++ b/olmo_core/model_and_optim/__3_10.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2e3b87bb5b741572b3ee0d7ea55da96d6520b7964d9f00d9afaafe399725fa75
+size 137667506
diff --git a/olmo_core/model_and_optim/__3_11.distcp b/olmo_core/model_and_optim/__3_11.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..ef19049920b21c27286fcf6a355d0cc1338eb974
--- /dev/null
+++ b/olmo_core/model_and_optim/__3_11.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dda1f63a04a45dcbab8b4eaf12ab598131a74d1f05ba71dd9cd4953bea90f36f
+size 137667506
diff --git a/olmo_core/model_and_optim/__3_12.distcp b/olmo_core/model_and_optim/__3_12.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..174228b418419e250f33ba56071a934c84d18ae2
--- /dev/null
+++ b/olmo_core/model_and_optim/__3_12.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:88f46e64feb162aaa12b0756d400d79a0d4daf634f3623fcbb3976c8537db6b8
+size 137676064
diff --git a/olmo_core/model_and_optim/__3_13.distcp b/olmo_core/model_and_optim/__3_13.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..d1545978e6da53ddbbd8e2b6e6c1982d0cb5909c
--- /dev/null
+++ b/olmo_core/model_and_optim/__3_13.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98a9cd09816e06cf8aa60adf88e30aabcca5e1a56fa6ceeb3e976e8624f439b5
+size 137676064
diff --git a/olmo_core/model_and_optim/__3_14.distcp b/olmo_core/model_and_optim/__3_14.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..a9c015c967940e5abbf252f738c98efaddb0c666
--- /dev/null
+++ b/olmo_core/model_and_optim/__3_14.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:50aaca15c3e0ec9bad026b4b32ee48ac36a80e2ddbca42a993c343de8ee7ce36
+size 137677641
diff --git a/olmo_core/model_and_optim/__3_15.distcp b/olmo_core/model_and_optim/__3_15.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..8b6a721a90b2fe6cd3847ce15d829833c58e3d24
--- /dev/null
+++ b/olmo_core/model_and_optim/__3_15.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2d8b403208455be48e7b6d424aacc488d391b3194d6ef422f289e18c134560d9
+size 137677641
diff --git a/olmo_core/model_and_optim/__3_2.distcp b/olmo_core/model_and_optim/__3_2.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..8eee2dfa2010e0f17b0f726b19a45014dc8980de
--- /dev/null
+++ b/olmo_core/model_and_optim/__3_2.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:091dfafb35f1d10c97ab35c6e9adf32b90012c7c0660d4377412acb9b154c5ac
+size 137975051
diff --git a/olmo_core/model_and_optim/__3_3.distcp b/olmo_core/model_and_optim/__3_3.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..3765b718193bd3ae7e7daa0a0dd55d2f4fa224d8
--- /dev/null
+++ b/olmo_core/model_and_optim/__3_3.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d90d5623b64e766c15c785b6e6f10b35933f5d68df40393390f4b2338de165d5
+size 138198775
diff --git a/olmo_core/model_and_optim/__3_4.distcp b/olmo_core/model_and_optim/__3_4.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..6f361053ad448072306b0bf1f8f3595043f728cb
--- /dev/null
+++ b/olmo_core/model_and_optim/__3_4.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0d06f99fa58634008f23265e4802df10f3350b33a2b5738131088a11649f85d7
+size 138070857
diff --git a/olmo_core/model_and_optim/__3_5.distcp b/olmo_core/model_and_optim/__3_5.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..3e4533cbe69f7508036ca028f6749dd5713dff46
--- /dev/null
+++ b/olmo_core/model_and_optim/__3_5.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3dfd90e58fdd2419a42a810181d088d508fe325cec62ad77c8b6f303fb58be66
+size 138070857
diff --git a/olmo_core/model_and_optim/__3_6.distcp b/olmo_core/model_and_optim/__3_6.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..a29b21a83603bf079c9c5b6119878c788085bf5a
--- /dev/null
+++ b/olmo_core/model_and_optim/__3_6.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:63b7a70571974f5e0ccd2be16cb67de1bdcd8d85c9947a7e09bfa2e76eb166e7
+size 138070857
diff --git a/olmo_core/model_and_optim/__3_7.distcp b/olmo_core/model_and_optim/__3_7.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..1b882cb3b9a8de836925515c607497aec1a11c23
--- /dev/null
+++ b/olmo_core/model_and_optim/__3_7.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:da8d1b72525528b6c6c388a117e89c2fc8479d046fd20ec149fe1f8a14abe89b
+size 137654337
diff --git a/olmo_core/model_and_optim/__3_8.distcp b/olmo_core/model_and_optim/__3_8.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..a1edbc3732db4c5a310249c070195f2cbe8496b3
--- /dev/null
+++ b/olmo_core/model_and_optim/__3_8.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3d6ae9940050e6f3effc633d254703fb5475bd6f20f0722df1f0735dbdddbf15
+size 137654337
diff --git a/olmo_core/model_and_optim/__3_9.distcp b/olmo_core/model_and_optim/__3_9.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..6daeac1b64ac4e3078439ee6cfb31cf86290feb9
--- /dev/null
+++ b/olmo_core/model_and_optim/__3_9.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:98d53f7740ce484023ec28ad52bc9e74fc01cb5f6d8fba6f58e2238a53234ab8
+size 137667506
diff --git a/olmo_core/model_and_optim/__4_0.distcp b/olmo_core/model_and_optim/__4_0.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..3bedd05aadf6e4b0a0c81dd8c342b2aead63240e
--- /dev/null
+++ b/olmo_core/model_and_optim/__4_0.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1b431577a0821391583f2c9f6b2a63de7229e2825cf3477b30560cced87abc2d
+size 137975051
diff --git a/olmo_core/model_and_optim/__4_1.distcp b/olmo_core/model_and_optim/__4_1.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..40f83690e87eb65f2416d76c24027062f5bfed49
--- /dev/null
+++ b/olmo_core/model_and_optim/__4_1.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:35f3ad093efb5870b5527656af6b57f9fbf5191cf0c7d3d1a02db87acbd00dfc
+size 137975051
diff --git a/olmo_core/model_and_optim/__4_10.distcp b/olmo_core/model_and_optim/__4_10.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..5e2848286b94947a2104c59a1c8209558c29abf1
--- /dev/null
+++ b/olmo_core/model_and_optim/__4_10.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:05ce569e25c40035dff2b127483e41d6569f24f16c2c51c9329170a962a419c1
+size 137667506
diff --git a/olmo_core/model_and_optim/__4_11.distcp b/olmo_core/model_and_optim/__4_11.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..6c17940f80e870425eb62d8143d957c8c2810391
--- /dev/null
+++ b/olmo_core/model_and_optim/__4_11.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e552ab1d30ae92add42f88438146b929129a3ddd81a069f527d9791a56d24e86
+size 137667506
diff --git a/olmo_core/model_and_optim/__4_12.distcp b/olmo_core/model_and_optim/__4_12.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..406127a3d455a2f607c8a395fe86f160c95b58a6
--- /dev/null
+++ b/olmo_core/model_and_optim/__4_12.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c1aaa87951508c67a7f01df43c812100273df8c71b80b651f9b46b19ca5c9aa
+size 137676064
diff --git a/olmo_core/model_and_optim/__4_13.distcp b/olmo_core/model_and_optim/__4_13.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..b90e2427065dde2edb43dc919a8b24bf1479f9fd
--- /dev/null
+++ b/olmo_core/model_and_optim/__4_13.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7b187e77aaff90f8506e4d0e022fa73494115aa76f1f6f431126204e4b9830e1
+size 137676064
diff --git a/olmo_core/model_and_optim/__4_14.distcp b/olmo_core/model_and_optim/__4_14.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..14a9c377e40a8b7e37f91851441653eb9cbb3b85
--- /dev/null
+++ b/olmo_core/model_and_optim/__4_14.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c20d514dc3e0303dcb54ce04f0418170b9744de70ea12c9d0962c469ccab773b
+size 137677641
diff --git a/olmo_core/model_and_optim/__4_15.distcp b/olmo_core/model_and_optim/__4_15.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..8258e16b3c6baab9a30e854a65076224a92515c9
--- /dev/null
+++ b/olmo_core/model_and_optim/__4_15.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d147e56945db6561a879e9a3fa9847e1714e76c6f800852b090b875a413f3266
+size 137677641
diff --git a/olmo_core/model_and_optim/__4_2.distcp b/olmo_core/model_and_optim/__4_2.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..48e681c4f63b0c28bd13667ce27cfa3bbe834fc6
--- /dev/null
+++ b/olmo_core/model_and_optim/__4_2.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:69bce1d162acbfc7e8d45abdde488496dd78cfcf0ec399c7cfd574c436844b57
+size 137975051
diff --git a/olmo_core/model_and_optim/__4_3.distcp b/olmo_core/model_and_optim/__4_3.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..b5ebbfa2cb8a25abf3cb2546db9d884be74267e7
--- /dev/null
+++ b/olmo_core/model_and_optim/__4_3.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b317390b29d2eda9d1cac883da6af812069f5f7fd7e5f5e59369c52417ec448b
+size 138198775
diff --git a/olmo_core/model_and_optim/__4_4.distcp b/olmo_core/model_and_optim/__4_4.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..b20bb785ba2b5b2f2757dc5704c338d0996e7fd8
--- /dev/null
+++ b/olmo_core/model_and_optim/__4_4.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2308054d7f4b399f87700035b631a200dae0b84dab0b918a1c0bab2777d54775
+size 138070857
diff --git a/olmo_core/model_and_optim/__4_5.distcp b/olmo_core/model_and_optim/__4_5.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..02683b5210ad7ad7b71240142670a4bb6eb08b77
--- /dev/null
+++ b/olmo_core/model_and_optim/__4_5.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:10c51b15344a53a9a1afaad0247ddb7368a64e1cce62601bf72ac8efd002b9b5
+size 138070857
diff --git a/olmo_core/model_and_optim/__4_6.distcp b/olmo_core/model_and_optim/__4_6.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..2a026894e9aab760191d8183bcb71112f40cb599
--- /dev/null
+++ b/olmo_core/model_and_optim/__4_6.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3dbefaf39e2c2dc270790413dc75757ad4f361b9372e41a6a5ce29806c4f1d99
+size 138070857
diff --git a/olmo_core/model_and_optim/__4_7.distcp b/olmo_core/model_and_optim/__4_7.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..c8e27bf7380b1c19d164432826b2ae0134f58995
--- /dev/null
+++ b/olmo_core/model_and_optim/__4_7.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3b9702586758150d68230e12793d9114eb7f26c7b6341eca89418f39b35ae6b3
+size 137654337
diff --git a/olmo_core/model_and_optim/__4_8.distcp b/olmo_core/model_and_optim/__4_8.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..0e954ae8bfc4dfa284e4c14e099997e08c0ab457
--- /dev/null
+++ b/olmo_core/model_and_optim/__4_8.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5c9d1d341fb9e47b3c88bef27f8c8e262fa470f30c11f708dea45636e1f9c9f3
+size 137654337
diff --git a/olmo_core/model_and_optim/__4_9.distcp b/olmo_core/model_and_optim/__4_9.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..814bbbe8fc1b2bd2f439c03d9f461787a6b81599
--- /dev/null
+++ b/olmo_core/model_and_optim/__4_9.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:7a63187dcef832f37626a4e92e63ab00068b35e1c054b9948495bb4f50e6a24f
+size 137667506
diff --git a/olmo_core/model_and_optim/__5_0.distcp b/olmo_core/model_and_optim/__5_0.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..f0ddb0a9cd86dd3ba256d73652f90f49a1174597
--- /dev/null
+++ b/olmo_core/model_and_optim/__5_0.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e44730a95b437c5ffb7785c1ede4e5dfe4c0381b0856a501d30b7272592222a9
+size 137975051
diff --git a/olmo_core/model_and_optim/__5_1.distcp b/olmo_core/model_and_optim/__5_1.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..98eb05d0581a2043025ae3ad343473f145b5b469
--- /dev/null
+++ b/olmo_core/model_and_optim/__5_1.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b7060744f51956571b4b87def256243b16ebaf621e6576d934e5b82af8ea232e
+size 137975051
diff --git a/olmo_core/model_and_optim/__5_10.distcp b/olmo_core/model_and_optim/__5_10.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..46eb9072668871dad91ec57040fd67af4ba8930c
--- /dev/null
+++ b/olmo_core/model_and_optim/__5_10.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:01293515e1a2227410bb3404a4e9110984d8504c716f2738498ff28cced97f6e
+size 137667506
diff --git a/olmo_core/model_and_optim/__5_11.distcp b/olmo_core/model_and_optim/__5_11.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..3d0215c85a08c95f6f2d79e7aebb3908b821f27c
--- /dev/null
+++ b/olmo_core/model_and_optim/__5_11.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b6287fd692ed09d53db9902d18c6cf3df0a03407df10a58f44a0064b13d97ea4
+size 137667506
diff --git a/olmo_core/model_and_optim/__5_12.distcp b/olmo_core/model_and_optim/__5_12.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..198bb8d923a0c9859cda84cb942c429e455252ee
--- /dev/null
+++ b/olmo_core/model_and_optim/__5_12.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:404fa5edb5e6665aa8a4dcca90b1c0db177aa2086648634818ab22a432b22f25
+size 137676064
diff --git a/olmo_core/model_and_optim/__5_13.distcp b/olmo_core/model_and_optim/__5_13.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..24fe91cb50d4e3de0c4ea60ca5eb442d0f4dc702
--- /dev/null
+++ b/olmo_core/model_and_optim/__5_13.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f2d9cb44ea2bd22dc458a5dd6e58b2b047dc3ee145f3ad416eeb4c39485b3588
+size 137676064
diff --git a/olmo_core/model_and_optim/__5_14.distcp b/olmo_core/model_and_optim/__5_14.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..130acf331be539c65b2fbf9ffd786b61dde0019e
--- /dev/null
+++ b/olmo_core/model_and_optim/__5_14.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9482d2599c78a282541839e36597767319c9348bed200b6e3c2f6371035b85da
+size 137677641
diff --git a/olmo_core/model_and_optim/__5_15.distcp b/olmo_core/model_and_optim/__5_15.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..5aaa0a5b68d2374ddae4241e21fc0f1223322956
--- /dev/null
+++ b/olmo_core/model_and_optim/__5_15.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4833bce320b4dd740814d7e1e1e5fa6f40f156ec9de6ff30d045dde09a05ac60
+size 137677641
diff --git a/olmo_core/model_and_optim/__5_2.distcp b/olmo_core/model_and_optim/__5_2.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..5ce8205a130a35eaddd0efeb4878856bfb15b61e
--- /dev/null
+++ b/olmo_core/model_and_optim/__5_2.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a0ea55fc7b9bd9fcb0016800048c83b5a51b53926bd40d69dd688604db239122
+size 137975051
diff --git a/olmo_core/model_and_optim/__5_3.distcp b/olmo_core/model_and_optim/__5_3.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..3d081093008bbcf8a92a1f8b10d9d5c95b30fe59
--- /dev/null
+++ b/olmo_core/model_and_optim/__5_3.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4f44989aa51b8a0bf5f46c06d6bec03098a13c491c1719d5887c69e99c6b0230
+size 138198775
diff --git a/olmo_core/model_and_optim/__5_4.distcp b/olmo_core/model_and_optim/__5_4.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..5b351436050f2c0da23228c980378bcd110cea5c
--- /dev/null
+++ b/olmo_core/model_and_optim/__5_4.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c010448227615a3517bc46a149093dd19d9997f95819b78ad645d50ae3025774
+size 138070857
diff --git a/olmo_core/model_and_optim/__5_5.distcp b/olmo_core/model_and_optim/__5_5.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..25dd7e6364a89d91429b378d9e45e823ad3bd7eb
--- /dev/null
+++ b/olmo_core/model_and_optim/__5_5.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:c0914cc1c6136828482331d6d7199096cc2c75d4f142296dce8a4cd54480e2ea
+size 138070857
diff --git a/olmo_core/model_and_optim/__5_6.distcp b/olmo_core/model_and_optim/__5_6.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..c587da62e5201bad5859521cd1b1d74bdabbb578
--- /dev/null
+++ b/olmo_core/model_and_optim/__5_6.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:f1dcb9aeed5dbdfb230bc820c4588969e746d9b23beb1fe4c457986f2e4d7b4c
+size 138070857
diff --git a/olmo_core/model_and_optim/__5_7.distcp b/olmo_core/model_and_optim/__5_7.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..b348e4f0512b7c8cdd75795c59d45fdccba838d4
--- /dev/null
+++ b/olmo_core/model_and_optim/__5_7.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6c59b70d0886f168407a7dd0e74432a3bfa5c113a809c6b70cab48f93065213e
+size 137654337
diff --git a/olmo_core/model_and_optim/__5_8.distcp b/olmo_core/model_and_optim/__5_8.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..4774db95e900674bcf255ae9981739286b870efa
--- /dev/null
+++ b/olmo_core/model_and_optim/__5_8.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:143e45a6523558305f7089faa3fbbb94c66b7b985be40c44de7600a73f59f22f
+size 137654337
diff --git a/olmo_core/model_and_optim/__5_9.distcp b/olmo_core/model_and_optim/__5_9.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..70a6af6a978f8390e8701330517cda8d5355adb1
--- /dev/null
+++ b/olmo_core/model_and_optim/__5_9.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef69c3b66a0333420a7348e47d2a5ca2382e1adceed1fb6a5cd038fcc6711281
+size 137667506
diff --git a/olmo_core/model_and_optim/__6_0.distcp b/olmo_core/model_and_optim/__6_0.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..70180b401d11346dba67aecb234f864b54fd7dfa
--- /dev/null
+++ b/olmo_core/model_and_optim/__6_0.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4631af3702ffdd19e1a59286b35ec4fb16fc6c4920110829663fec90419b1d87
+size 137975051
diff --git a/olmo_core/model_and_optim/__6_1.distcp b/olmo_core/model_and_optim/__6_1.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..13430fee5fe023eff835dc9e1558123cbd3bf36c
--- /dev/null
+++ b/olmo_core/model_and_optim/__6_1.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:b0c99411c8ddde3d93c083eaabdfb7b7fcfbc17eabdf645dbef049cef5c20906
+size 137975051
diff --git a/olmo_core/model_and_optim/__6_10.distcp b/olmo_core/model_and_optim/__6_10.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..8f01686646cea2e1d0b6f237d1b46f51baee5246
--- /dev/null
+++ b/olmo_core/model_and_optim/__6_10.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:aa268dc3afdc8c9cdd411a5f0236f308431040b9721a3cda9cc304f3845a058b
+size 137667506
diff --git a/olmo_core/model_and_optim/__6_11.distcp b/olmo_core/model_and_optim/__6_11.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..edbede24d6b8f2385dc79e3039746802470bb0c0
--- /dev/null
+++ b/olmo_core/model_and_optim/__6_11.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d89110046268dd79cb08c473630b759e646a92d7f682c165d4ee97c4982faab4
+size 137667506
diff --git a/olmo_core/model_and_optim/__6_12.distcp b/olmo_core/model_and_optim/__6_12.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..17053cba6e745ab452028e2d79c690f0dc945738
--- /dev/null
+++ b/olmo_core/model_and_optim/__6_12.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8ba139450c8057148335dfd65db889d3d8344813df9bab91233ac9b92a514e64
+size 137676064
diff --git a/olmo_core/model_and_optim/__6_13.distcp b/olmo_core/model_and_optim/__6_13.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..aecfccdb9c625b4cc6996ae4d0bca456b514f9a5
--- /dev/null
+++ b/olmo_core/model_and_optim/__6_13.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:87b820a18ba2379cd500e5d9488ba14961de7c4a89f708707c1f089cf8b4c31c
+size 137676064
diff --git a/olmo_core/model_and_optim/__6_14.distcp b/olmo_core/model_and_optim/__6_14.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..6a0bcfb963ef6336b204fbedccde462496f32fe3
--- /dev/null
+++ b/olmo_core/model_and_optim/__6_14.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fbc08ad6036f804df725fae59e206babae8f1eafcc8a98516659b4692e91de5a
+size 137677641
diff --git a/olmo_core/model_and_optim/__6_15.distcp b/olmo_core/model_and_optim/__6_15.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..1f22f5265017b3b9fab832d7a262be09f11a5de5
--- /dev/null
+++ b/olmo_core/model_and_optim/__6_15.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0c60497f9023f67308fec3aefa174c3ebc9a80dc0b7c31e3d16407ee4c36f5b0
+size 137677641
diff --git a/olmo_core/model_and_optim/__6_2.distcp b/olmo_core/model_and_optim/__6_2.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..4e426b7bda9c4d3e57efc883a9308c32fd74ef37
--- /dev/null
+++ b/olmo_core/model_and_optim/__6_2.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:174dd6404393b7a149d8551ba85f699ade73830ad546387f1fdf09af025374e1
+size 137975051
diff --git a/olmo_core/model_and_optim/__6_3.distcp b/olmo_core/model_and_optim/__6_3.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..8b2acd5c469446ee300785a5f1c1ab69e3529435
--- /dev/null
+++ b/olmo_core/model_and_optim/__6_3.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:14152cba746c9f4198fd97dc2c85d9a87a8e1d6501825571814da43d054f0f02
+size 138198775
diff --git a/olmo_core/model_and_optim/__6_4.distcp b/olmo_core/model_and_optim/__6_4.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..6a63507bb25c712890ba2f6a1a309fde391bed3e
--- /dev/null
+++ b/olmo_core/model_and_optim/__6_4.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:39f3b01a57275cf9ca90a778310a886075ba24c533f2be28b32d194d53c4a6cd
+size 138070857
diff --git a/olmo_core/model_and_optim/__6_5.distcp b/olmo_core/model_and_optim/__6_5.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..ad5cf579c0b0867bd3ebf3b64424f3c1d4d5d606
--- /dev/null
+++ b/olmo_core/model_and_optim/__6_5.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:a9733b9025077901405b1fe0e47a165ce09b218b543012731fe3bfda2e7beb2f
+size 138070857
diff --git a/olmo_core/model_and_optim/__6_6.distcp b/olmo_core/model_and_optim/__6_6.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..fbc05f805245bfcfdf4a2102e46f7617d1873df1
--- /dev/null
+++ b/olmo_core/model_and_optim/__6_6.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e478df6a79a98efb77b9d6a76520e62de395d3dac5697e9ffcb620f592fa4e0b
+size 138070857
diff --git a/olmo_core/model_and_optim/__6_7.distcp b/olmo_core/model_and_optim/__6_7.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..a727fabbe0e9156ae35010a44e35c2a26c9dd4dd
--- /dev/null
+++ b/olmo_core/model_and_optim/__6_7.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:67aa4a690d3966adc2573ae639c593ba6e1e06a0fef7e0a39dec434a44448bca
+size 137654337
diff --git a/olmo_core/model_and_optim/__6_8.distcp b/olmo_core/model_and_optim/__6_8.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..93f167c64572d26cefcf2507ce31706df26762cd
--- /dev/null
+++ b/olmo_core/model_and_optim/__6_8.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3980a27c98ffe56e3863cd666039a347294ba600b24bd31a2804a97c1a8173c2
+size 137654337
diff --git a/olmo_core/model_and_optim/__6_9.distcp b/olmo_core/model_and_optim/__6_9.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..b3c4cda55dfcf41935f19c7f783924bb35519eca
--- /dev/null
+++ b/olmo_core/model_and_optim/__6_9.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:2c09fb86474e20be21bca35066115831769a2d6cbf74bcbe4fe0c6606385ac8c
+size 137667506
diff --git a/olmo_core/model_and_optim/__7_0.distcp b/olmo_core/model_and_optim/__7_0.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..db3d4e69d3ed46c5aacec018d1bba7c479990c97
--- /dev/null
+++ b/olmo_core/model_and_optim/__7_0.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:61896453db219bb860f6b3f0f5ba07a61aed2ccf88bf8a4c13772a90a6da3fe4
+size 137958667
diff --git a/olmo_core/model_and_optim/__7_1.distcp b/olmo_core/model_and_optim/__7_1.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..d64f8afcab60a36868a96f187489324f86e3115f
--- /dev/null
+++ b/olmo_core/model_and_optim/__7_1.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1ff5c8249cbcd180326285a0fa5622c4817915200f93841d1094bb1908361f98
+size 137958667
diff --git a/olmo_core/model_and_optim/__7_10.distcp b/olmo_core/model_and_optim/__7_10.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..4803941ffb61c84200bedcbd5ebc1d0a0bdcdb55
--- /dev/null
+++ b/olmo_core/model_and_optim/__7_10.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:54c2f3a9c0f11308214c3b9a70f24b64e173c386ebcc3b023eb93cbd241930a1
+size 137667506
diff --git a/olmo_core/model_and_optim/__7_11.distcp b/olmo_core/model_and_optim/__7_11.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..6f6c0c7a002e98e5d123558059e9da8488b69025
--- /dev/null
+++ b/olmo_core/model_and_optim/__7_11.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:66d544fd4348795804369c4af14015219697b4029dbbaf0e227bb068055ad9ae
+size 137667506
diff --git a/olmo_core/model_and_optim/__7_12.distcp b/olmo_core/model_and_optim/__7_12.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..8d9584de44cf7b42d3cade47c2e9b986384764e3
--- /dev/null
+++ b/olmo_core/model_and_optim/__7_12.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:9690c8068efce2675387eff5da27164cc6409348fc0ecc3079ca061bc518a3f6
+size 137676064
diff --git a/olmo_core/model_and_optim/__7_13.distcp b/olmo_core/model_and_optim/__7_13.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..68986121fa904e78a87b479451b98df579c5980b
--- /dev/null
+++ b/olmo_core/model_and_optim/__7_13.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0f1ce5b3118c722a49586082328d488ac98c1fcd041a28124c84d6b7c9b5448b
+size 137676064
diff --git a/olmo_core/model_and_optim/__7_14.distcp b/olmo_core/model_and_optim/__7_14.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..00c780b40e6fdbdf073b03e845f3aa26609e837f
--- /dev/null
+++ b/olmo_core/model_and_optim/__7_14.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e2ea242adadf9a77058b3cdbdc9d8dbd6c954885638844821e0feace1ac9a6bf
+size 137677641
diff --git a/olmo_core/model_and_optim/__7_15.distcp b/olmo_core/model_and_optim/__7_15.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..a63f690df1650eeef4d72f3646a4843a88459f35
--- /dev/null
+++ b/olmo_core/model_and_optim/__7_15.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:0eb8619f7e99de9a7710e07de48965ea91f76ff535a24494f2eb1e1d652b9d09
+size 137677641
diff --git a/olmo_core/model_and_optim/__7_2.distcp b/olmo_core/model_and_optim/__7_2.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..1851b333f71bb0a5a6cd4f8a4b302d5afb13b52a
--- /dev/null
+++ b/olmo_core/model_and_optim/__7_2.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e201525a561a809a3808480cf120624af54cfaf56d6f23114cea91af22924e75
+size 137958667
diff --git a/olmo_core/model_and_optim/__7_3.distcp b/olmo_core/model_and_optim/__7_3.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..b498c206342c4c0d81bae9d1eced1b66b74ac216
--- /dev/null
+++ b/olmo_core/model_and_optim/__7_3.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:1f33231175907855b3ec0734ef818b1d159cb22084b6c5ef613b4fcae73bc0ef
+size 138198775
diff --git a/olmo_core/model_and_optim/__7_4.distcp b/olmo_core/model_and_optim/__7_4.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..823b8ea8726005abdcf40ba9135f652db6813784
--- /dev/null
+++ b/olmo_core/model_and_optim/__7_4.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6a8bfb5c7fe3fc7cc01d0d0ce7661a213183731b724e5e2fdba75243a0439742
+size 138070857
diff --git a/olmo_core/model_and_optim/__7_5.distcp b/olmo_core/model_and_optim/__7_5.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..625894d9c4aff3ba8ce6ffef3a1907bf8e09e331
--- /dev/null
+++ b/olmo_core/model_and_optim/__7_5.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:6d9f5d967da28d259fa1eb706ca138d4c447012d6aae81563ebb19ee59d30a6f
+size 138070857
diff --git a/olmo_core/model_and_optim/__7_6.distcp b/olmo_core/model_and_optim/__7_6.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..c3613bc617240612db66184d2c0b259b01579f53
--- /dev/null
+++ b/olmo_core/model_and_optim/__7_6.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:719fefa07cf5eec54e3a5608463b596fd25588084d4ffa2ac9c13ae44bcd34ee
+size 138070857
diff --git a/olmo_core/model_and_optim/__7_7.distcp b/olmo_core/model_and_optim/__7_7.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..571cca712b368b1c6851cbc38f447a873381c029
--- /dev/null
+++ b/olmo_core/model_and_optim/__7_7.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:e726167a0a0357819d1f2c9ac2412fa05aa9b8cd25da397c2edba98d62028978
+size 137654337
diff --git a/olmo_core/model_and_optim/__7_8.distcp b/olmo_core/model_and_optim/__7_8.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..b1b4d21b0df4f292f7620a85e778f5e3ee6d3ba6
--- /dev/null
+++ b/olmo_core/model_and_optim/__7_8.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d8277e71d22723632e0224e86e2299f7e7b0f24db7f79c17a05b8938ec8a56a9
+size 137654337
diff --git a/olmo_core/model_and_optim/__7_9.distcp b/olmo_core/model_and_optim/__7_9.distcp
new file mode 100644
index 0000000000000000000000000000000000000000..93cdf371426def412fe29d4cfe9c27d720775f5c
--- /dev/null
+++ b/olmo_core/model_and_optim/__7_9.distcp
@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:4180bec34e0d2835b304f52e968824246a8c6e37bf0b2fefcd1c48024a0c29d6
+size 137667506