Ibbyml's picture
Upload folder using huggingface_hub
5bc4606 verified
Raw History Blame Contribute Delete
27.2 kB
{"array_metadatas": [{"array_metadata": {"param_name": "tok.emb", "write_shape": [204800, 2048], "chunk_shape": [51200, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "tok.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.0.attn.q", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.0.attn.k", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.0.attn.v", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.0.attn.out", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.0.attn.sinks", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.0.attn.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.0.attn.qnorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.0.attn.knorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.0.attn.sdpa_gate", "write_shape": [2048, 16], "chunk_shape": [2048, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.0.mlp.mlp_up", "write_shape": [2048, 2, 3072], "chunk_shape": [2048, 2, 3072], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.0.mlp.mlp_down", "write_shape": [3072, 2048], "chunk_shape": [3072, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.0.mlp.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.1.attn.q", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.1.attn.k", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.1.attn.v", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.1.attn.out", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.1.attn.sinks", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.1.attn.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.1.attn.qnorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.1.attn.knorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.1.attn.sdpa_gate", "write_shape": [2048, 16], "chunk_shape": [2048, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.1.mlp.mlp_up", "write_shape": [2048, 2, 3072], "chunk_shape": [2048, 2, 3072], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.1.mlp.mlp_down", "write_shape": [3072, 2048], "chunk_shape": [3072, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.1.mlp.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.2.attn.q", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.2.attn.k", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.2.attn.v", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.2.attn.out", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.2.attn.sinks", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.2.attn.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.2.attn.qnorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.2.attn.knorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.2.attn.sdpa_gate", "write_shape": [2048, 16], "chunk_shape": [2048, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.2.mlp.mlp_up", "write_shape": [2048, 2, 3072], "chunk_shape": [2048, 2, 3072], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.2.mlp.mlp_down", "write_shape": [3072, 2048], "chunk_shape": [3072, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.2.mlp.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.3.attn.q", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.3.attn.k", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.3.attn.v", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.3.attn.out", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.3.attn.sinks", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.3.attn.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.3.attn.qnorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.3.attn.knorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.3.attn.sdpa_gate", "write_shape": [2048, 16], "chunk_shape": [2048, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.3.mlp.router", "write_shape": [2048, 15], "chunk_shape": [2048, 15], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.3.mlp.router_bias", "write_shape": [15], "chunk_shape": [15], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.3.mlp.mlp_up", "write_shape": [16, 2048, 2, 2048], "chunk_shape": [16, 1024, 2, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.3.mlp.mlp_down", "write_shape": [16, 2048, 2048], "chunk_shape": [16, 2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.3.mlp.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.4.attn.q", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.4.attn.k", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.4.attn.v", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.4.attn.out", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.4.attn.sinks", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.4.attn.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.4.attn.qnorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.4.attn.knorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.4.attn.sdpa_gate", "write_shape": [2048, 16], "chunk_shape": [2048, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.4.mlp.mlp_up", "write_shape": [2048, 2, 3072], "chunk_shape": [2048, 2, 3072], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.4.mlp.mlp_down", "write_shape": [3072, 2048], "chunk_shape": [3072, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.4.mlp.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.5.attn.q", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.5.attn.k", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.5.attn.v", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.5.attn.out", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.5.attn.sinks", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.5.attn.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.5.attn.qnorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.5.attn.knorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.5.attn.sdpa_gate", "write_shape": [2048, 16], "chunk_shape": [2048, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.5.mlp.mlp_up", "write_shape": [2048, 2, 3072], "chunk_shape": [2048, 2, 3072], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.5.mlp.mlp_down", "write_shape": [3072, 2048], "chunk_shape": [3072, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.5.mlp.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.6.attn.q", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.6.attn.k", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.6.attn.v", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.6.attn.out", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.6.attn.sinks", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.6.attn.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.6.attn.qnorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.6.attn.knorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.6.attn.sdpa_gate", "write_shape": [2048, 16], "chunk_shape": [2048, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.6.mlp.mlp_up", "write_shape": [2048, 2, 3072], "chunk_shape": [2048, 2, 3072], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.6.mlp.mlp_down", "write_shape": [3072, 2048], "chunk_shape": [3072, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.6.mlp.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.7.attn.q", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.7.attn.k", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.7.attn.v", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.7.attn.out", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.7.attn.sinks", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.7.attn.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.7.attn.qnorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.7.attn.knorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.7.attn.sdpa_gate", "write_shape": [2048, 16], "chunk_shape": [2048, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.7.mlp.router", "write_shape": [2048, 15], "chunk_shape": [2048, 15], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.7.mlp.router_bias", "write_shape": [15], "chunk_shape": [15], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.7.mlp.mlp_up", "write_shape": [16, 2048, 2, 2048], "chunk_shape": [16, 1024, 2, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.7.mlp.mlp_down", "write_shape": [16, 2048, 2048], "chunk_shape": [16, 2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.7.mlp.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.8.attn.q", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.8.attn.k", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.8.attn.v", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.8.attn.out", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.8.attn.sinks", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.8.attn.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.8.attn.qnorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.8.attn.knorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.8.attn.sdpa_gate", "write_shape": [2048, 16], "chunk_shape": [2048, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.8.mlp.mlp_up", "write_shape": [2048, 2, 3072], "chunk_shape": [2048, 2, 3072], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.8.mlp.mlp_down", "write_shape": [3072, 2048], "chunk_shape": [3072, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.8.mlp.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.9.attn.q", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.9.attn.k", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.9.attn.v", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.9.attn.out", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.9.attn.sinks", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.9.attn.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.9.attn.qnorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.9.attn.knorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.9.attn.sdpa_gate", "write_shape": [2048, 16], "chunk_shape": [2048, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.9.mlp.mlp_up", "write_shape": [2048, 2, 3072], "chunk_shape": [2048, 2, 3072], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.9.mlp.mlp_down", "write_shape": [3072, 2048], "chunk_shape": [3072, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.9.mlp.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.10.attn.q", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.10.attn.k", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.10.attn.v", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.10.attn.out", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.10.attn.sinks", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.10.attn.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.10.attn.qnorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.10.attn.knorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.10.attn.sdpa_gate", "write_shape": [2048, 16], "chunk_shape": [2048, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.10.mlp.mlp_up", "write_shape": [2048, 2, 3072], "chunk_shape": [2048, 2, 3072], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.10.mlp.mlp_down", "write_shape": [3072, 2048], "chunk_shape": [3072, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.10.mlp.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.11.attn.q", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.11.attn.k", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.11.attn.v", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.11.attn.out", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.11.attn.sinks", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.11.attn.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.11.attn.qnorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.11.attn.knorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.11.attn.sdpa_gate", "write_shape": [2048, 16], "chunk_shape": [2048, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.11.mlp.router", "write_shape": [2048, 15], "chunk_shape": [2048, 15], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.11.mlp.router_bias", "write_shape": [15], "chunk_shape": [15], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.11.mlp.mlp_up", "write_shape": [16, 2048, 2, 2048], "chunk_shape": [16, 1024, 2, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.11.mlp.mlp_down", "write_shape": [16, 2048, 2048], "chunk_shape": [16, 2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.11.mlp.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.12.attn.q", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.12.attn.k", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.12.attn.v", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.12.attn.out", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.12.attn.sinks", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.12.attn.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.12.attn.qnorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.12.attn.knorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.12.attn.sdpa_gate", "write_shape": [2048, 16], "chunk_shape": [2048, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.12.mlp.mlp_up", "write_shape": [2048, 2, 3072], "chunk_shape": [2048, 2, 3072], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.12.mlp.mlp_down", "write_shape": [3072, 2048], "chunk_shape": [3072, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.12.mlp.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.13.attn.q", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.13.attn.k", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.13.attn.v", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.13.attn.out", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.13.attn.sinks", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.13.attn.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.13.attn.qnorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.13.attn.knorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.13.attn.sdpa_gate", "write_shape": [2048, 16], "chunk_shape": [2048, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.13.mlp.mlp_up", "write_shape": [2048, 2, 3072], "chunk_shape": [2048, 2, 3072], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.13.mlp.mlp_down", "write_shape": [3072, 2048], "chunk_shape": [3072, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.13.mlp.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.14.attn.q", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.14.attn.k", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.14.attn.v", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.14.attn.out", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.14.attn.sinks", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.14.attn.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.14.attn.qnorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.14.attn.knorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.14.attn.sdpa_gate", "write_shape": [2048, 16], "chunk_shape": [2048, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.14.mlp.mlp_up", "write_shape": [2048, 2, 3072], "chunk_shape": [2048, 2, 3072], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.14.mlp.mlp_down", "write_shape": [3072, 2048], "chunk_shape": [3072, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.14.mlp.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.15.attn.q", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.15.attn.k", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.15.attn.v", "write_shape": [2048, 512], "chunk_shape": [2048, 512], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.15.attn.out", "write_shape": [2048, 2048], "chunk_shape": [2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.15.attn.sinks", "write_shape": [16], "chunk_shape": [16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.15.attn.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.15.attn.qnorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.15.attn.knorm_scale", "write_shape": [128], "chunk_shape": [128], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.15.attn.sdpa_gate", "write_shape": [2048, 16], "chunk_shape": [2048, 16], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.15.mlp.router", "write_shape": [2048, 15], "chunk_shape": [2048, 15], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.15.mlp.router_bias", "write_shape": [15], "chunk_shape": [15], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.15.mlp.mlp_up", "write_shape": [16, 2048, 2, 2048], "chunk_shape": [16, 1024, 2, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.15.mlp.mlp_down", "write_shape": [16, 2048, 2048], "chunk_shape": [16, 2048, 2048], "ext_metadata": null}}, {"array_metadata": {"param_name": "layers.15.mlp.norm_scale", "write_shape": [2048], "chunk_shape": [2048], "ext_metadata": null}}]}