Add files using upload-large-folder tool

.gitattributes

@@ -44,3 +44,7 @@ sft_checkpoints/state.zarr/.mu\[2\].mlp.c_proj/c/1/2 filter=lfs diff=lfs merge=l
 sft_checkpoints/state.zarr/.mu\[2\].mlp.c_proj/c/3/0 filter=lfs diff=lfs merge=lfs -text
 sft_checkpoints/state.zarr/.mu\[2\].mlp.c_proj/c/3/1 filter=lfs diff=lfs merge=lfs -text
 sft_checkpoints/state.zarr/.mu\[2\].mlp.c_proj/c/3/3 filter=lfs diff=lfs merge=lfs -text
+sft_checkpoints/state.zarr/.mu\[2\].mlp.c_proj/c/3/2 filter=lfs diff=lfs merge=lfs -text
+sft_checkpoints/state.zarr/.mu\[2\].mlp.c_proj/c/2/0 filter=lfs diff=lfs merge=lfs -text
+sft_checkpoints/state.zarr/.mu\[2\].mlp.c_proj/c/2/1 filter=lfs diff=lfs merge=lfs -text
+sft_checkpoints/state.zarr/.mu\[2\].mlp.c_proj/c/2/3 filter=lfs diff=lfs merge=lfs -text

base_checkpoints/model.zarr/.h[0].attn.c_proj/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

base_checkpoints/model.zarr/.h[10].mlp.c_proj/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    3072,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        768,
+        192
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

base_checkpoints/model.zarr/.h[1].mlp.c_fc/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    3072
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        768
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

base_checkpoints/model.zarr/.h[3].mlp.c_proj/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    3072,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        768,
+        192
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

base_checkpoints/model.zarr/.h[5].attn.c_q/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

base_checkpoints/model.zarr/.h[5].attn.c_v/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

base_checkpoints/model.zarr/.h[6].mlp.c_proj/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    3072,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        768,
+        192
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

base_checkpoints/model.zarr/.h[7].attn.c_proj/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

base_checkpoints/model.zarr/.h[8].attn.c_k/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

base_checkpoints/model.zarr/.h[8].attn.c_q/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

base_checkpoints/model.zarr/.h[8].attn.c_v/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

base_checkpoints/model.zarr/.h[9].mlp.c_proj/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    3072,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        768,
+        192
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/model.zarr/zarr.json

@@ -0,0 +1,14 @@
+{
+  "attributes": {
+    "config": {
+      "sequence_len": 1024,
+      "vocab_size": 32768,
+      "n_layer": 12,
+      "n_head": 3,
+      "n_kv_head": 3,
+      "n_embed": 768
+    }
+  },
+  "zarr_format": 3,
+  "node_type": "group"
+}

dpo_checkpoints/state.zarr/.adamw_mu[1]/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    32768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        48,
+        4096
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.adamw_nu[1]/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    32768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        48,
+        4096
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[0].attn.c_k/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[0].attn.c_proj/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[0].attn.c_q/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[0].attn.c_v/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[0].mlp.c_fc/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    3072
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        768
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[0].mlp.c_proj/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    3072,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        768,
+        192
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[10].attn.c_k/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[10].attn.c_q/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[10].attn.c_v/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[10].mlp.c_proj/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    3072,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        768,
+        192
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[11].attn.c_k/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[11].attn.c_q/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[11].attn.c_v/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[11].mlp.c_fc/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    3072
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        768
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[11].mlp.c_proj/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    3072,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        768,
+        192
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[1].attn.c_k/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[1].attn.c_q/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[1].attn.c_v/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[1].mlp.c_proj/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    3072,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        768,
+        192
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[2].attn.c_k/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[2].attn.c_proj/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[2].attn.c_q/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[2].attn.c_v/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[2].mlp.c_fc/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    3072
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        768
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[3].attn.c_proj/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[3].mlp.c_fc/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    3072
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        768
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[3].mlp.c_proj/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    3072,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        768,
+        192
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[4].attn.c_k/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[4].attn.c_proj/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[4].attn.c_q/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[4].attn.c_v/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[4].mlp.c_fc/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    3072
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        768
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[4].mlp.c_proj/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    3072,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        768,
+        192
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}

dpo_checkpoints/state.zarr/.mu[5].attn.c_proj/zarr.json

@@ -0,0 +1,42 @@
+{
+  "shape": [
+    768,
+    768
+  ],
+  "data_type": "float32",
+  "chunk_grid": {
+    "name": "regular",
+    "configuration": {
+      "chunk_shape": [
+        192,
+        384
+      ]
+    }
+  },
+  "chunk_key_encoding": {
+    "name": "default",
+    "configuration": {
+      "separator": "/"
+    }
+  },
+  "fill_value": 0.0,
+  "codecs": [
+    {
+      "name": "bytes",
+      "configuration": {
+        "endian": "little"
+      }
+    },
+    {
+      "name": "zstd",
+      "configuration": {
+        "level": 0,
+        "checksum": false
+      }
+    }
+  ],
+  "attributes": {},
+  "zarr_format": 3,
+  "node_type": "array",
+  "storage_transformers": []
+}