Download model_structure.json from EasyDeL/gpt-oss-20b: direct link, hf CLI and curl.
- Browser
- Download file 94.6 kB
-
https://huggingface.co/EasyDeL/gpt-oss-20b/resolve/main/model_structure.json
- Command line
-
hf download hf://EasyDeL/gpt-oss-20b/model_structure.json
-
curl -L -o model_structure.json https://huggingface.co/EasyDeL/gpt-oss-20b/resolve/main/model_structure.json
94.6 kB
| { | |
| "format": "pytree-structure", | |
| "version": "0.0.84", | |
| "backend": "tensorstore", | |
| "prefix": "model", | |
| "treedef_b64": "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", | |
| "leaf_keys_full": [ | |
| "model.lm_head.kernel", | |
| "model.model.embed_tokens.embedding", | |
| "model.model.layers.0.input_layernorm.kernel", | |
| "model.model.layers.0.mlp.experts.down_proj.bias", | |
| "model.model.layers.0.mlp.experts.down_proj.kernel", | |
| "model.model.layers.0.mlp.experts.gate_proj.bias", | |
| "model.model.layers.0.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.0.mlp.experts.up_proj.bias", | |
| "model.model.layers.0.mlp.experts.up_proj.kernel", | |
| "model.model.layers.0.mlp.router.bias", | |
| "model.model.layers.0.mlp.router.kernel", | |
| "model.model.layers.0.post_attention_layernorm.kernel", | |
| "model.model.layers.0.self_attn.k_proj.bias", | |
| "model.model.layers.0.self_attn.k_proj.kernel", | |
| "model.model.layers.0.self_attn.o_proj.bias", | |
| "model.model.layers.0.self_attn.o_proj.kernel", | |
| "model.model.layers.0.self_attn.q_proj.bias", | |
| "model.model.layers.0.self_attn.q_proj.kernel", | |
| "model.model.layers.0.self_attn.sinks", | |
| "model.model.layers.0.self_attn.v_proj.bias", | |
| "model.model.layers.0.self_attn.v_proj.kernel", | |
| "model.model.layers.1.input_layernorm.kernel", | |
| "model.model.layers.1.mlp.experts.down_proj.bias", | |
| "model.model.layers.1.mlp.experts.down_proj.kernel", | |
| "model.model.layers.1.mlp.experts.gate_proj.bias", | |
| "model.model.layers.1.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.1.mlp.experts.up_proj.bias", | |
| "model.model.layers.1.mlp.experts.up_proj.kernel", | |
| "model.model.layers.1.mlp.router.bias", | |
| "model.model.layers.1.mlp.router.kernel", | |
| "model.model.layers.1.post_attention_layernorm.kernel", | |
| "model.model.layers.1.self_attn.k_proj.bias", | |
| "model.model.layers.1.self_attn.k_proj.kernel", | |
| "model.model.layers.1.self_attn.o_proj.bias", | |
| "model.model.layers.1.self_attn.o_proj.kernel", | |
| "model.model.layers.1.self_attn.q_proj.bias", | |
| "model.model.layers.1.self_attn.q_proj.kernel", | |
| "model.model.layers.1.self_attn.sinks", | |
| "model.model.layers.1.self_attn.v_proj.bias", | |
| "model.model.layers.1.self_attn.v_proj.kernel", | |
| "model.model.layers.2.input_layernorm.kernel", | |
| "model.model.layers.2.mlp.experts.down_proj.bias", | |
| "model.model.layers.2.mlp.experts.down_proj.kernel", | |
| "model.model.layers.2.mlp.experts.gate_proj.bias", | |
| "model.model.layers.2.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.2.mlp.experts.up_proj.bias", | |
| "model.model.layers.2.mlp.experts.up_proj.kernel", | |
| "model.model.layers.2.mlp.router.bias", | |
| "model.model.layers.2.mlp.router.kernel", | |
| "model.model.layers.2.post_attention_layernorm.kernel", | |
| "model.model.layers.2.self_attn.k_proj.bias", | |
| "model.model.layers.2.self_attn.k_proj.kernel", | |
| "model.model.layers.2.self_attn.o_proj.bias", | |
| "model.model.layers.2.self_attn.o_proj.kernel", | |
| "model.model.layers.2.self_attn.q_proj.bias", | |
| "model.model.layers.2.self_attn.q_proj.kernel", | |
| "model.model.layers.2.self_attn.sinks", | |
| "model.model.layers.2.self_attn.v_proj.bias", | |
| "model.model.layers.2.self_attn.v_proj.kernel", | |
| "model.model.layers.3.input_layernorm.kernel", | |
| "model.model.layers.3.mlp.experts.down_proj.bias", | |
| "model.model.layers.3.mlp.experts.down_proj.kernel", | |
| "model.model.layers.3.mlp.experts.gate_proj.bias", | |
| "model.model.layers.3.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.3.mlp.experts.up_proj.bias", | |
| "model.model.layers.3.mlp.experts.up_proj.kernel", | |
| "model.model.layers.3.mlp.router.bias", | |
| "model.model.layers.3.mlp.router.kernel", | |
| "model.model.layers.3.post_attention_layernorm.kernel", | |
| "model.model.layers.3.self_attn.k_proj.bias", | |
| "model.model.layers.3.self_attn.k_proj.kernel", | |
| "model.model.layers.3.self_attn.o_proj.bias", | |
| "model.model.layers.3.self_attn.o_proj.kernel", | |
| "model.model.layers.3.self_attn.q_proj.bias", | |
| "model.model.layers.3.self_attn.q_proj.kernel", | |
| "model.model.layers.3.self_attn.sinks", | |
| "model.model.layers.3.self_attn.v_proj.bias", | |
| "model.model.layers.3.self_attn.v_proj.kernel", | |
| "model.model.layers.4.input_layernorm.kernel", | |
| "model.model.layers.4.mlp.experts.down_proj.bias", | |
| "model.model.layers.4.mlp.experts.down_proj.kernel", | |
| "model.model.layers.4.mlp.experts.gate_proj.bias", | |
| "model.model.layers.4.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.4.mlp.experts.up_proj.bias", | |
| "model.model.layers.4.mlp.experts.up_proj.kernel", | |
| "model.model.layers.4.mlp.router.bias", | |
| "model.model.layers.4.mlp.router.kernel", | |
| "model.model.layers.4.post_attention_layernorm.kernel", | |
| "model.model.layers.4.self_attn.k_proj.bias", | |
| "model.model.layers.4.self_attn.k_proj.kernel", | |
| "model.model.layers.4.self_attn.o_proj.bias", | |
| "model.model.layers.4.self_attn.o_proj.kernel", | |
| "model.model.layers.4.self_attn.q_proj.bias", | |
| "model.model.layers.4.self_attn.q_proj.kernel", | |
| "model.model.layers.4.self_attn.sinks", | |
| "model.model.layers.4.self_attn.v_proj.bias", | |
| "model.model.layers.4.self_attn.v_proj.kernel", | |
| "model.model.layers.5.input_layernorm.kernel", | |
| "model.model.layers.5.mlp.experts.down_proj.bias", | |
| "model.model.layers.5.mlp.experts.down_proj.kernel", | |
| "model.model.layers.5.mlp.experts.gate_proj.bias", | |
| "model.model.layers.5.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.5.mlp.experts.up_proj.bias", | |
| "model.model.layers.5.mlp.experts.up_proj.kernel", | |
| "model.model.layers.5.mlp.router.bias", | |
| "model.model.layers.5.mlp.router.kernel", | |
| "model.model.layers.5.post_attention_layernorm.kernel", | |
| "model.model.layers.5.self_attn.k_proj.bias", | |
| "model.model.layers.5.self_attn.k_proj.kernel", | |
| "model.model.layers.5.self_attn.o_proj.bias", | |
| "model.model.layers.5.self_attn.o_proj.kernel", | |
| "model.model.layers.5.self_attn.q_proj.bias", | |
| "model.model.layers.5.self_attn.q_proj.kernel", | |
| "model.model.layers.5.self_attn.sinks", | |
| "model.model.layers.5.self_attn.v_proj.bias", | |
| "model.model.layers.5.self_attn.v_proj.kernel", | |
| "model.model.layers.6.input_layernorm.kernel", | |
| "model.model.layers.6.mlp.experts.down_proj.bias", | |
| "model.model.layers.6.mlp.experts.down_proj.kernel", | |
| "model.model.layers.6.mlp.experts.gate_proj.bias", | |
| "model.model.layers.6.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.6.mlp.experts.up_proj.bias", | |
| "model.model.layers.6.mlp.experts.up_proj.kernel", | |
| "model.model.layers.6.mlp.router.bias", | |
| "model.model.layers.6.mlp.router.kernel", | |
| "model.model.layers.6.post_attention_layernorm.kernel", | |
| "model.model.layers.6.self_attn.k_proj.bias", | |
| "model.model.layers.6.self_attn.k_proj.kernel", | |
| "model.model.layers.6.self_attn.o_proj.bias", | |
| "model.model.layers.6.self_attn.o_proj.kernel", | |
| "model.model.layers.6.self_attn.q_proj.bias", | |
| "model.model.layers.6.self_attn.q_proj.kernel", | |
| "model.model.layers.6.self_attn.sinks", | |
| "model.model.layers.6.self_attn.v_proj.bias", | |
| "model.model.layers.6.self_attn.v_proj.kernel", | |
| "model.model.layers.7.input_layernorm.kernel", | |
| "model.model.layers.7.mlp.experts.down_proj.bias", | |
| "model.model.layers.7.mlp.experts.down_proj.kernel", | |
| "model.model.layers.7.mlp.experts.gate_proj.bias", | |
| "model.model.layers.7.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.7.mlp.experts.up_proj.bias", | |
| "model.model.layers.7.mlp.experts.up_proj.kernel", | |
| "model.model.layers.7.mlp.router.bias", | |
| "model.model.layers.7.mlp.router.kernel", | |
| "model.model.layers.7.post_attention_layernorm.kernel", | |
| "model.model.layers.7.self_attn.k_proj.bias", | |
| "model.model.layers.7.self_attn.k_proj.kernel", | |
| "model.model.layers.7.self_attn.o_proj.bias", | |
| "model.model.layers.7.self_attn.o_proj.kernel", | |
| "model.model.layers.7.self_attn.q_proj.bias", | |
| "model.model.layers.7.self_attn.q_proj.kernel", | |
| "model.model.layers.7.self_attn.sinks", | |
| "model.model.layers.7.self_attn.v_proj.bias", | |
| "model.model.layers.7.self_attn.v_proj.kernel", | |
| "model.model.layers.8.input_layernorm.kernel", | |
| "model.model.layers.8.mlp.experts.down_proj.bias", | |
| "model.model.layers.8.mlp.experts.down_proj.kernel", | |
| "model.model.layers.8.mlp.experts.gate_proj.bias", | |
| "model.model.layers.8.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.8.mlp.experts.up_proj.bias", | |
| "model.model.layers.8.mlp.experts.up_proj.kernel", | |
| "model.model.layers.8.mlp.router.bias", | |
| "model.model.layers.8.mlp.router.kernel", | |
| "model.model.layers.8.post_attention_layernorm.kernel", | |
| "model.model.layers.8.self_attn.k_proj.bias", | |
| "model.model.layers.8.self_attn.k_proj.kernel", | |
| "model.model.layers.8.self_attn.o_proj.bias", | |
| "model.model.layers.8.self_attn.o_proj.kernel", | |
| "model.model.layers.8.self_attn.q_proj.bias", | |
| "model.model.layers.8.self_attn.q_proj.kernel", | |
| "model.model.layers.8.self_attn.sinks", | |
| "model.model.layers.8.self_attn.v_proj.bias", | |
| "model.model.layers.8.self_attn.v_proj.kernel", | |
| "model.model.layers.9.input_layernorm.kernel", | |
| "model.model.layers.9.mlp.experts.down_proj.bias", | |
| "model.model.layers.9.mlp.experts.down_proj.kernel", | |
| "model.model.layers.9.mlp.experts.gate_proj.bias", | |
| "model.model.layers.9.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.9.mlp.experts.up_proj.bias", | |
| "model.model.layers.9.mlp.experts.up_proj.kernel", | |
| "model.model.layers.9.mlp.router.bias", | |
| "model.model.layers.9.mlp.router.kernel", | |
| "model.model.layers.9.post_attention_layernorm.kernel", | |
| "model.model.layers.9.self_attn.k_proj.bias", | |
| "model.model.layers.9.self_attn.k_proj.kernel", | |
| "model.model.layers.9.self_attn.o_proj.bias", | |
| "model.model.layers.9.self_attn.o_proj.kernel", | |
| "model.model.layers.9.self_attn.q_proj.bias", | |
| "model.model.layers.9.self_attn.q_proj.kernel", | |
| "model.model.layers.9.self_attn.sinks", | |
| "model.model.layers.9.self_attn.v_proj.bias", | |
| "model.model.layers.9.self_attn.v_proj.kernel", | |
| "model.model.layers.10.input_layernorm.kernel", | |
| "model.model.layers.10.mlp.experts.down_proj.bias", | |
| "model.model.layers.10.mlp.experts.down_proj.kernel", | |
| "model.model.layers.10.mlp.experts.gate_proj.bias", | |
| "model.model.layers.10.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.10.mlp.experts.up_proj.bias", | |
| "model.model.layers.10.mlp.experts.up_proj.kernel", | |
| "model.model.layers.10.mlp.router.bias", | |
| "model.model.layers.10.mlp.router.kernel", | |
| "model.model.layers.10.post_attention_layernorm.kernel", | |
| "model.model.layers.10.self_attn.k_proj.bias", | |
| "model.model.layers.10.self_attn.k_proj.kernel", | |
| "model.model.layers.10.self_attn.o_proj.bias", | |
| "model.model.layers.10.self_attn.o_proj.kernel", | |
| "model.model.layers.10.self_attn.q_proj.bias", | |
| "model.model.layers.10.self_attn.q_proj.kernel", | |
| "model.model.layers.10.self_attn.sinks", | |
| "model.model.layers.10.self_attn.v_proj.bias", | |
| "model.model.layers.10.self_attn.v_proj.kernel", | |
| "model.model.layers.11.input_layernorm.kernel", | |
| "model.model.layers.11.mlp.experts.down_proj.bias", | |
| "model.model.layers.11.mlp.experts.down_proj.kernel", | |
| "model.model.layers.11.mlp.experts.gate_proj.bias", | |
| "model.model.layers.11.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.11.mlp.experts.up_proj.bias", | |
| "model.model.layers.11.mlp.experts.up_proj.kernel", | |
| "model.model.layers.11.mlp.router.bias", | |
| "model.model.layers.11.mlp.router.kernel", | |
| "model.model.layers.11.post_attention_layernorm.kernel", | |
| "model.model.layers.11.self_attn.k_proj.bias", | |
| "model.model.layers.11.self_attn.k_proj.kernel", | |
| "model.model.layers.11.self_attn.o_proj.bias", | |
| "model.model.layers.11.self_attn.o_proj.kernel", | |
| "model.model.layers.11.self_attn.q_proj.bias", | |
| "model.model.layers.11.self_attn.q_proj.kernel", | |
| "model.model.layers.11.self_attn.sinks", | |
| "model.model.layers.11.self_attn.v_proj.bias", | |
| "model.model.layers.11.self_attn.v_proj.kernel", | |
| "model.model.layers.12.input_layernorm.kernel", | |
| "model.model.layers.12.mlp.experts.down_proj.bias", | |
| "model.model.layers.12.mlp.experts.down_proj.kernel", | |
| "model.model.layers.12.mlp.experts.gate_proj.bias", | |
| "model.model.layers.12.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.12.mlp.experts.up_proj.bias", | |
| "model.model.layers.12.mlp.experts.up_proj.kernel", | |
| "model.model.layers.12.mlp.router.bias", | |
| "model.model.layers.12.mlp.router.kernel", | |
| "model.model.layers.12.post_attention_layernorm.kernel", | |
| "model.model.layers.12.self_attn.k_proj.bias", | |
| "model.model.layers.12.self_attn.k_proj.kernel", | |
| "model.model.layers.12.self_attn.o_proj.bias", | |
| "model.model.layers.12.self_attn.o_proj.kernel", | |
| "model.model.layers.12.self_attn.q_proj.bias", | |
| "model.model.layers.12.self_attn.q_proj.kernel", | |
| "model.model.layers.12.self_attn.sinks", | |
| "model.model.layers.12.self_attn.v_proj.bias", | |
| "model.model.layers.12.self_attn.v_proj.kernel", | |
| "model.model.layers.13.input_layernorm.kernel", | |
| "model.model.layers.13.mlp.experts.down_proj.bias", | |
| "model.model.layers.13.mlp.experts.down_proj.kernel", | |
| "model.model.layers.13.mlp.experts.gate_proj.bias", | |
| "model.model.layers.13.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.13.mlp.experts.up_proj.bias", | |
| "model.model.layers.13.mlp.experts.up_proj.kernel", | |
| "model.model.layers.13.mlp.router.bias", | |
| "model.model.layers.13.mlp.router.kernel", | |
| "model.model.layers.13.post_attention_layernorm.kernel", | |
| "model.model.layers.13.self_attn.k_proj.bias", | |
| "model.model.layers.13.self_attn.k_proj.kernel", | |
| "model.model.layers.13.self_attn.o_proj.bias", | |
| "model.model.layers.13.self_attn.o_proj.kernel", | |
| "model.model.layers.13.self_attn.q_proj.bias", | |
| "model.model.layers.13.self_attn.q_proj.kernel", | |
| "model.model.layers.13.self_attn.sinks", | |
| "model.model.layers.13.self_attn.v_proj.bias", | |
| "model.model.layers.13.self_attn.v_proj.kernel", | |
| "model.model.layers.14.input_layernorm.kernel", | |
| "model.model.layers.14.mlp.experts.down_proj.bias", | |
| "model.model.layers.14.mlp.experts.down_proj.kernel", | |
| "model.model.layers.14.mlp.experts.gate_proj.bias", | |
| "model.model.layers.14.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.14.mlp.experts.up_proj.bias", | |
| "model.model.layers.14.mlp.experts.up_proj.kernel", | |
| "model.model.layers.14.mlp.router.bias", | |
| "model.model.layers.14.mlp.router.kernel", | |
| "model.model.layers.14.post_attention_layernorm.kernel", | |
| "model.model.layers.14.self_attn.k_proj.bias", | |
| "model.model.layers.14.self_attn.k_proj.kernel", | |
| "model.model.layers.14.self_attn.o_proj.bias", | |
| "model.model.layers.14.self_attn.o_proj.kernel", | |
| "model.model.layers.14.self_attn.q_proj.bias", | |
| "model.model.layers.14.self_attn.q_proj.kernel", | |
| "model.model.layers.14.self_attn.sinks", | |
| "model.model.layers.14.self_attn.v_proj.bias", | |
| "model.model.layers.14.self_attn.v_proj.kernel", | |
| "model.model.layers.15.input_layernorm.kernel", | |
| "model.model.layers.15.mlp.experts.down_proj.bias", | |
| "model.model.layers.15.mlp.experts.down_proj.kernel", | |
| "model.model.layers.15.mlp.experts.gate_proj.bias", | |
| "model.model.layers.15.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.15.mlp.experts.up_proj.bias", | |
| "model.model.layers.15.mlp.experts.up_proj.kernel", | |
| "model.model.layers.15.mlp.router.bias", | |
| "model.model.layers.15.mlp.router.kernel", | |
| "model.model.layers.15.post_attention_layernorm.kernel", | |
| "model.model.layers.15.self_attn.k_proj.bias", | |
| "model.model.layers.15.self_attn.k_proj.kernel", | |
| "model.model.layers.15.self_attn.o_proj.bias", | |
| "model.model.layers.15.self_attn.o_proj.kernel", | |
| "model.model.layers.15.self_attn.q_proj.bias", | |
| "model.model.layers.15.self_attn.q_proj.kernel", | |
| "model.model.layers.15.self_attn.sinks", | |
| "model.model.layers.15.self_attn.v_proj.bias", | |
| "model.model.layers.15.self_attn.v_proj.kernel", | |
| "model.model.layers.16.input_layernorm.kernel", | |
| "model.model.layers.16.mlp.experts.down_proj.bias", | |
| "model.model.layers.16.mlp.experts.down_proj.kernel", | |
| "model.model.layers.16.mlp.experts.gate_proj.bias", | |
| "model.model.layers.16.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.16.mlp.experts.up_proj.bias", | |
| "model.model.layers.16.mlp.experts.up_proj.kernel", | |
| "model.model.layers.16.mlp.router.bias", | |
| "model.model.layers.16.mlp.router.kernel", | |
| "model.model.layers.16.post_attention_layernorm.kernel", | |
| "model.model.layers.16.self_attn.k_proj.bias", | |
| "model.model.layers.16.self_attn.k_proj.kernel", | |
| "model.model.layers.16.self_attn.o_proj.bias", | |
| "model.model.layers.16.self_attn.o_proj.kernel", | |
| "model.model.layers.16.self_attn.q_proj.bias", | |
| "model.model.layers.16.self_attn.q_proj.kernel", | |
| "model.model.layers.16.self_attn.sinks", | |
| "model.model.layers.16.self_attn.v_proj.bias", | |
| "model.model.layers.16.self_attn.v_proj.kernel", | |
| "model.model.layers.17.input_layernorm.kernel", | |
| "model.model.layers.17.mlp.experts.down_proj.bias", | |
| "model.model.layers.17.mlp.experts.down_proj.kernel", | |
| "model.model.layers.17.mlp.experts.gate_proj.bias", | |
| "model.model.layers.17.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.17.mlp.experts.up_proj.bias", | |
| "model.model.layers.17.mlp.experts.up_proj.kernel", | |
| "model.model.layers.17.mlp.router.bias", | |
| "model.model.layers.17.mlp.router.kernel", | |
| "model.model.layers.17.post_attention_layernorm.kernel", | |
| "model.model.layers.17.self_attn.k_proj.bias", | |
| "model.model.layers.17.self_attn.k_proj.kernel", | |
| "model.model.layers.17.self_attn.o_proj.bias", | |
| "model.model.layers.17.self_attn.o_proj.kernel", | |
| "model.model.layers.17.self_attn.q_proj.bias", | |
| "model.model.layers.17.self_attn.q_proj.kernel", | |
| "model.model.layers.17.self_attn.sinks", | |
| "model.model.layers.17.self_attn.v_proj.bias", | |
| "model.model.layers.17.self_attn.v_proj.kernel", | |
| "model.model.layers.18.input_layernorm.kernel", | |
| "model.model.layers.18.mlp.experts.down_proj.bias", | |
| "model.model.layers.18.mlp.experts.down_proj.kernel", | |
| "model.model.layers.18.mlp.experts.gate_proj.bias", | |
| "model.model.layers.18.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.18.mlp.experts.up_proj.bias", | |
| "model.model.layers.18.mlp.experts.up_proj.kernel", | |
| "model.model.layers.18.mlp.router.bias", | |
| "model.model.layers.18.mlp.router.kernel", | |
| "model.model.layers.18.post_attention_layernorm.kernel", | |
| "model.model.layers.18.self_attn.k_proj.bias", | |
| "model.model.layers.18.self_attn.k_proj.kernel", | |
| "model.model.layers.18.self_attn.o_proj.bias", | |
| "model.model.layers.18.self_attn.o_proj.kernel", | |
| "model.model.layers.18.self_attn.q_proj.bias", | |
| "model.model.layers.18.self_attn.q_proj.kernel", | |
| "model.model.layers.18.self_attn.sinks", | |
| "model.model.layers.18.self_attn.v_proj.bias", | |
| "model.model.layers.18.self_attn.v_proj.kernel", | |
| "model.model.layers.19.input_layernorm.kernel", | |
| "model.model.layers.19.mlp.experts.down_proj.bias", | |
| "model.model.layers.19.mlp.experts.down_proj.kernel", | |
| "model.model.layers.19.mlp.experts.gate_proj.bias", | |
| "model.model.layers.19.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.19.mlp.experts.up_proj.bias", | |
| "model.model.layers.19.mlp.experts.up_proj.kernel", | |
| "model.model.layers.19.mlp.router.bias", | |
| "model.model.layers.19.mlp.router.kernel", | |
| "model.model.layers.19.post_attention_layernorm.kernel", | |
| "model.model.layers.19.self_attn.k_proj.bias", | |
| "model.model.layers.19.self_attn.k_proj.kernel", | |
| "model.model.layers.19.self_attn.o_proj.bias", | |
| "model.model.layers.19.self_attn.o_proj.kernel", | |
| "model.model.layers.19.self_attn.q_proj.bias", | |
| "model.model.layers.19.self_attn.q_proj.kernel", | |
| "model.model.layers.19.self_attn.sinks", | |
| "model.model.layers.19.self_attn.v_proj.bias", | |
| "model.model.layers.19.self_attn.v_proj.kernel", | |
| "model.model.layers.20.input_layernorm.kernel", | |
| "model.model.layers.20.mlp.experts.down_proj.bias", | |
| "model.model.layers.20.mlp.experts.down_proj.kernel", | |
| "model.model.layers.20.mlp.experts.gate_proj.bias", | |
| "model.model.layers.20.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.20.mlp.experts.up_proj.bias", | |
| "model.model.layers.20.mlp.experts.up_proj.kernel", | |
| "model.model.layers.20.mlp.router.bias", | |
| "model.model.layers.20.mlp.router.kernel", | |
| "model.model.layers.20.post_attention_layernorm.kernel", | |
| "model.model.layers.20.self_attn.k_proj.bias", | |
| "model.model.layers.20.self_attn.k_proj.kernel", | |
| "model.model.layers.20.self_attn.o_proj.bias", | |
| "model.model.layers.20.self_attn.o_proj.kernel", | |
| "model.model.layers.20.self_attn.q_proj.bias", | |
| "model.model.layers.20.self_attn.q_proj.kernel", | |
| "model.model.layers.20.self_attn.sinks", | |
| "model.model.layers.20.self_attn.v_proj.bias", | |
| "model.model.layers.20.self_attn.v_proj.kernel", | |
| "model.model.layers.21.input_layernorm.kernel", | |
| "model.model.layers.21.mlp.experts.down_proj.bias", | |
| "model.model.layers.21.mlp.experts.down_proj.kernel", | |
| "model.model.layers.21.mlp.experts.gate_proj.bias", | |
| "model.model.layers.21.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.21.mlp.experts.up_proj.bias", | |
| "model.model.layers.21.mlp.experts.up_proj.kernel", | |
| "model.model.layers.21.mlp.router.bias", | |
| "model.model.layers.21.mlp.router.kernel", | |
| "model.model.layers.21.post_attention_layernorm.kernel", | |
| "model.model.layers.21.self_attn.k_proj.bias", | |
| "model.model.layers.21.self_attn.k_proj.kernel", | |
| "model.model.layers.21.self_attn.o_proj.bias", | |
| "model.model.layers.21.self_attn.o_proj.kernel", | |
| "model.model.layers.21.self_attn.q_proj.bias", | |
| "model.model.layers.21.self_attn.q_proj.kernel", | |
| "model.model.layers.21.self_attn.sinks", | |
| "model.model.layers.21.self_attn.v_proj.bias", | |
| "model.model.layers.21.self_attn.v_proj.kernel", | |
| "model.model.layers.22.input_layernorm.kernel", | |
| "model.model.layers.22.mlp.experts.down_proj.bias", | |
| "model.model.layers.22.mlp.experts.down_proj.kernel", | |
| "model.model.layers.22.mlp.experts.gate_proj.bias", | |
| "model.model.layers.22.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.22.mlp.experts.up_proj.bias", | |
| "model.model.layers.22.mlp.experts.up_proj.kernel", | |
| "model.model.layers.22.mlp.router.bias", | |
| "model.model.layers.22.mlp.router.kernel", | |
| "model.model.layers.22.post_attention_layernorm.kernel", | |
| "model.model.layers.22.self_attn.k_proj.bias", | |
| "model.model.layers.22.self_attn.k_proj.kernel", | |
| "model.model.layers.22.self_attn.o_proj.bias", | |
| "model.model.layers.22.self_attn.o_proj.kernel", | |
| "model.model.layers.22.self_attn.q_proj.bias", | |
| "model.model.layers.22.self_attn.q_proj.kernel", | |
| "model.model.layers.22.self_attn.sinks", | |
| "model.model.layers.22.self_attn.v_proj.bias", | |
| "model.model.layers.22.self_attn.v_proj.kernel", | |
| "model.model.layers.23.input_layernorm.kernel", | |
| "model.model.layers.23.mlp.experts.down_proj.bias", | |
| "model.model.layers.23.mlp.experts.down_proj.kernel", | |
| "model.model.layers.23.mlp.experts.gate_proj.bias", | |
| "model.model.layers.23.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.23.mlp.experts.up_proj.bias", | |
| "model.model.layers.23.mlp.experts.up_proj.kernel", | |
| "model.model.layers.23.mlp.router.bias", | |
| "model.model.layers.23.mlp.router.kernel", | |
| "model.model.layers.23.post_attention_layernorm.kernel", | |
| "model.model.layers.23.self_attn.k_proj.bias", | |
| "model.model.layers.23.self_attn.k_proj.kernel", | |
| "model.model.layers.23.self_attn.o_proj.bias", | |
| "model.model.layers.23.self_attn.o_proj.kernel", | |
| "model.model.layers.23.self_attn.q_proj.bias", | |
| "model.model.layers.23.self_attn.q_proj.kernel", | |
| "model.model.layers.23.self_attn.sinks", | |
| "model.model.layers.23.self_attn.v_proj.bias", | |
| "model.model.layers.23.self_attn.v_proj.kernel", | |
| "model.model.norm.kernel" | |
| ], | |
| "arr_mask": [ | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true, | |
| true | |
| ], | |
| "array_keys": [ | |
| "model.lm_head.kernel", | |
| "model.model.embed_tokens.embedding", | |
| "model.model.layers.0.input_layernorm.kernel", | |
| "model.model.layers.0.mlp.experts.down_proj.bias", | |
| "model.model.layers.0.mlp.experts.down_proj.kernel", | |
| "model.model.layers.0.mlp.experts.gate_proj.bias", | |
| "model.model.layers.0.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.0.mlp.experts.up_proj.bias", | |
| "model.model.layers.0.mlp.experts.up_proj.kernel", | |
| "model.model.layers.0.mlp.router.bias", | |
| "model.model.layers.0.mlp.router.kernel", | |
| "model.model.layers.0.post_attention_layernorm.kernel", | |
| "model.model.layers.0.self_attn.k_proj.bias", | |
| "model.model.layers.0.self_attn.k_proj.kernel", | |
| "model.model.layers.0.self_attn.o_proj.bias", | |
| "model.model.layers.0.self_attn.o_proj.kernel", | |
| "model.model.layers.0.self_attn.q_proj.bias", | |
| "model.model.layers.0.self_attn.q_proj.kernel", | |
| "model.model.layers.0.self_attn.sinks", | |
| "model.model.layers.0.self_attn.v_proj.bias", | |
| "model.model.layers.0.self_attn.v_proj.kernel", | |
| "model.model.layers.1.input_layernorm.kernel", | |
| "model.model.layers.1.mlp.experts.down_proj.bias", | |
| "model.model.layers.1.mlp.experts.down_proj.kernel", | |
| "model.model.layers.1.mlp.experts.gate_proj.bias", | |
| "model.model.layers.1.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.1.mlp.experts.up_proj.bias", | |
| "model.model.layers.1.mlp.experts.up_proj.kernel", | |
| "model.model.layers.1.mlp.router.bias", | |
| "model.model.layers.1.mlp.router.kernel", | |
| "model.model.layers.1.post_attention_layernorm.kernel", | |
| "model.model.layers.1.self_attn.k_proj.bias", | |
| "model.model.layers.1.self_attn.k_proj.kernel", | |
| "model.model.layers.1.self_attn.o_proj.bias", | |
| "model.model.layers.1.self_attn.o_proj.kernel", | |
| "model.model.layers.1.self_attn.q_proj.bias", | |
| "model.model.layers.1.self_attn.q_proj.kernel", | |
| "model.model.layers.1.self_attn.sinks", | |
| "model.model.layers.1.self_attn.v_proj.bias", | |
| "model.model.layers.1.self_attn.v_proj.kernel", | |
| "model.model.layers.2.input_layernorm.kernel", | |
| "model.model.layers.2.mlp.experts.down_proj.bias", | |
| "model.model.layers.2.mlp.experts.down_proj.kernel", | |
| "model.model.layers.2.mlp.experts.gate_proj.bias", | |
| "model.model.layers.2.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.2.mlp.experts.up_proj.bias", | |
| "model.model.layers.2.mlp.experts.up_proj.kernel", | |
| "model.model.layers.2.mlp.router.bias", | |
| "model.model.layers.2.mlp.router.kernel", | |
| "model.model.layers.2.post_attention_layernorm.kernel", | |
| "model.model.layers.2.self_attn.k_proj.bias", | |
| "model.model.layers.2.self_attn.k_proj.kernel", | |
| "model.model.layers.2.self_attn.o_proj.bias", | |
| "model.model.layers.2.self_attn.o_proj.kernel", | |
| "model.model.layers.2.self_attn.q_proj.bias", | |
| "model.model.layers.2.self_attn.q_proj.kernel", | |
| "model.model.layers.2.self_attn.sinks", | |
| "model.model.layers.2.self_attn.v_proj.bias", | |
| "model.model.layers.2.self_attn.v_proj.kernel", | |
| "model.model.layers.3.input_layernorm.kernel", | |
| "model.model.layers.3.mlp.experts.down_proj.bias", | |
| "model.model.layers.3.mlp.experts.down_proj.kernel", | |
| "model.model.layers.3.mlp.experts.gate_proj.bias", | |
| "model.model.layers.3.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.3.mlp.experts.up_proj.bias", | |
| "model.model.layers.3.mlp.experts.up_proj.kernel", | |
| "model.model.layers.3.mlp.router.bias", | |
| "model.model.layers.3.mlp.router.kernel", | |
| "model.model.layers.3.post_attention_layernorm.kernel", | |
| "model.model.layers.3.self_attn.k_proj.bias", | |
| "model.model.layers.3.self_attn.k_proj.kernel", | |
| "model.model.layers.3.self_attn.o_proj.bias", | |
| "model.model.layers.3.self_attn.o_proj.kernel", | |
| "model.model.layers.3.self_attn.q_proj.bias", | |
| "model.model.layers.3.self_attn.q_proj.kernel", | |
| "model.model.layers.3.self_attn.sinks", | |
| "model.model.layers.3.self_attn.v_proj.bias", | |
| "model.model.layers.3.self_attn.v_proj.kernel", | |
| "model.model.layers.4.input_layernorm.kernel", | |
| "model.model.layers.4.mlp.experts.down_proj.bias", | |
| "model.model.layers.4.mlp.experts.down_proj.kernel", | |
| "model.model.layers.4.mlp.experts.gate_proj.bias", | |
| "model.model.layers.4.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.4.mlp.experts.up_proj.bias", | |
| "model.model.layers.4.mlp.experts.up_proj.kernel", | |
| "model.model.layers.4.mlp.router.bias", | |
| "model.model.layers.4.mlp.router.kernel", | |
| "model.model.layers.4.post_attention_layernorm.kernel", | |
| "model.model.layers.4.self_attn.k_proj.bias", | |
| "model.model.layers.4.self_attn.k_proj.kernel", | |
| "model.model.layers.4.self_attn.o_proj.bias", | |
| "model.model.layers.4.self_attn.o_proj.kernel", | |
| "model.model.layers.4.self_attn.q_proj.bias", | |
| "model.model.layers.4.self_attn.q_proj.kernel", | |
| "model.model.layers.4.self_attn.sinks", | |
| "model.model.layers.4.self_attn.v_proj.bias", | |
| "model.model.layers.4.self_attn.v_proj.kernel", | |
| "model.model.layers.5.input_layernorm.kernel", | |
| "model.model.layers.5.mlp.experts.down_proj.bias", | |
| "model.model.layers.5.mlp.experts.down_proj.kernel", | |
| "model.model.layers.5.mlp.experts.gate_proj.bias", | |
| "model.model.layers.5.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.5.mlp.experts.up_proj.bias", | |
| "model.model.layers.5.mlp.experts.up_proj.kernel", | |
| "model.model.layers.5.mlp.router.bias", | |
| "model.model.layers.5.mlp.router.kernel", | |
| "model.model.layers.5.post_attention_layernorm.kernel", | |
| "model.model.layers.5.self_attn.k_proj.bias", | |
| "model.model.layers.5.self_attn.k_proj.kernel", | |
| "model.model.layers.5.self_attn.o_proj.bias", | |
| "model.model.layers.5.self_attn.o_proj.kernel", | |
| "model.model.layers.5.self_attn.q_proj.bias", | |
| "model.model.layers.5.self_attn.q_proj.kernel", | |
| "model.model.layers.5.self_attn.sinks", | |
| "model.model.layers.5.self_attn.v_proj.bias", | |
| "model.model.layers.5.self_attn.v_proj.kernel", | |
| "model.model.layers.6.input_layernorm.kernel", | |
| "model.model.layers.6.mlp.experts.down_proj.bias", | |
| "model.model.layers.6.mlp.experts.down_proj.kernel", | |
| "model.model.layers.6.mlp.experts.gate_proj.bias", | |
| "model.model.layers.6.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.6.mlp.experts.up_proj.bias", | |
| "model.model.layers.6.mlp.experts.up_proj.kernel", | |
| "model.model.layers.6.mlp.router.bias", | |
| "model.model.layers.6.mlp.router.kernel", | |
| "model.model.layers.6.post_attention_layernorm.kernel", | |
| "model.model.layers.6.self_attn.k_proj.bias", | |
| "model.model.layers.6.self_attn.k_proj.kernel", | |
| "model.model.layers.6.self_attn.o_proj.bias", | |
| "model.model.layers.6.self_attn.o_proj.kernel", | |
| "model.model.layers.6.self_attn.q_proj.bias", | |
| "model.model.layers.6.self_attn.q_proj.kernel", | |
| "model.model.layers.6.self_attn.sinks", | |
| "model.model.layers.6.self_attn.v_proj.bias", | |
| "model.model.layers.6.self_attn.v_proj.kernel", | |
| "model.model.layers.7.input_layernorm.kernel", | |
| "model.model.layers.7.mlp.experts.down_proj.bias", | |
| "model.model.layers.7.mlp.experts.down_proj.kernel", | |
| "model.model.layers.7.mlp.experts.gate_proj.bias", | |
| "model.model.layers.7.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.7.mlp.experts.up_proj.bias", | |
| "model.model.layers.7.mlp.experts.up_proj.kernel", | |
| "model.model.layers.7.mlp.router.bias", | |
| "model.model.layers.7.mlp.router.kernel", | |
| "model.model.layers.7.post_attention_layernorm.kernel", | |
| "model.model.layers.7.self_attn.k_proj.bias", | |
| "model.model.layers.7.self_attn.k_proj.kernel", | |
| "model.model.layers.7.self_attn.o_proj.bias", | |
| "model.model.layers.7.self_attn.o_proj.kernel", | |
| "model.model.layers.7.self_attn.q_proj.bias", | |
| "model.model.layers.7.self_attn.q_proj.kernel", | |
| "model.model.layers.7.self_attn.sinks", | |
| "model.model.layers.7.self_attn.v_proj.bias", | |
| "model.model.layers.7.self_attn.v_proj.kernel", | |
| "model.model.layers.8.input_layernorm.kernel", | |
| "model.model.layers.8.mlp.experts.down_proj.bias", | |
| "model.model.layers.8.mlp.experts.down_proj.kernel", | |
| "model.model.layers.8.mlp.experts.gate_proj.bias", | |
| "model.model.layers.8.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.8.mlp.experts.up_proj.bias", | |
| "model.model.layers.8.mlp.experts.up_proj.kernel", | |
| "model.model.layers.8.mlp.router.bias", | |
| "model.model.layers.8.mlp.router.kernel", | |
| "model.model.layers.8.post_attention_layernorm.kernel", | |
| "model.model.layers.8.self_attn.k_proj.bias", | |
| "model.model.layers.8.self_attn.k_proj.kernel", | |
| "model.model.layers.8.self_attn.o_proj.bias", | |
| "model.model.layers.8.self_attn.o_proj.kernel", | |
| "model.model.layers.8.self_attn.q_proj.bias", | |
| "model.model.layers.8.self_attn.q_proj.kernel", | |
| "model.model.layers.8.self_attn.sinks", | |
| "model.model.layers.8.self_attn.v_proj.bias", | |
| "model.model.layers.8.self_attn.v_proj.kernel", | |
| "model.model.layers.9.input_layernorm.kernel", | |
| "model.model.layers.9.mlp.experts.down_proj.bias", | |
| "model.model.layers.9.mlp.experts.down_proj.kernel", | |
| "model.model.layers.9.mlp.experts.gate_proj.bias", | |
| "model.model.layers.9.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.9.mlp.experts.up_proj.bias", | |
| "model.model.layers.9.mlp.experts.up_proj.kernel", | |
| "model.model.layers.9.mlp.router.bias", | |
| "model.model.layers.9.mlp.router.kernel", | |
| "model.model.layers.9.post_attention_layernorm.kernel", | |
| "model.model.layers.9.self_attn.k_proj.bias", | |
| "model.model.layers.9.self_attn.k_proj.kernel", | |
| "model.model.layers.9.self_attn.o_proj.bias", | |
| "model.model.layers.9.self_attn.o_proj.kernel", | |
| "model.model.layers.9.self_attn.q_proj.bias", | |
| "model.model.layers.9.self_attn.q_proj.kernel", | |
| "model.model.layers.9.self_attn.sinks", | |
| "model.model.layers.9.self_attn.v_proj.bias", | |
| "model.model.layers.9.self_attn.v_proj.kernel", | |
| "model.model.layers.10.input_layernorm.kernel", | |
| "model.model.layers.10.mlp.experts.down_proj.bias", | |
| "model.model.layers.10.mlp.experts.down_proj.kernel", | |
| "model.model.layers.10.mlp.experts.gate_proj.bias", | |
| "model.model.layers.10.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.10.mlp.experts.up_proj.bias", | |
| "model.model.layers.10.mlp.experts.up_proj.kernel", | |
| "model.model.layers.10.mlp.router.bias", | |
| "model.model.layers.10.mlp.router.kernel", | |
| "model.model.layers.10.post_attention_layernorm.kernel", | |
| "model.model.layers.10.self_attn.k_proj.bias", | |
| "model.model.layers.10.self_attn.k_proj.kernel", | |
| "model.model.layers.10.self_attn.o_proj.bias", | |
| "model.model.layers.10.self_attn.o_proj.kernel", | |
| "model.model.layers.10.self_attn.q_proj.bias", | |
| "model.model.layers.10.self_attn.q_proj.kernel", | |
| "model.model.layers.10.self_attn.sinks", | |
| "model.model.layers.10.self_attn.v_proj.bias", | |
| "model.model.layers.10.self_attn.v_proj.kernel", | |
| "model.model.layers.11.input_layernorm.kernel", | |
| "model.model.layers.11.mlp.experts.down_proj.bias", | |
| "model.model.layers.11.mlp.experts.down_proj.kernel", | |
| "model.model.layers.11.mlp.experts.gate_proj.bias", | |
| "model.model.layers.11.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.11.mlp.experts.up_proj.bias", | |
| "model.model.layers.11.mlp.experts.up_proj.kernel", | |
| "model.model.layers.11.mlp.router.bias", | |
| "model.model.layers.11.mlp.router.kernel", | |
| "model.model.layers.11.post_attention_layernorm.kernel", | |
| "model.model.layers.11.self_attn.k_proj.bias", | |
| "model.model.layers.11.self_attn.k_proj.kernel", | |
| "model.model.layers.11.self_attn.o_proj.bias", | |
| "model.model.layers.11.self_attn.o_proj.kernel", | |
| "model.model.layers.11.self_attn.q_proj.bias", | |
| "model.model.layers.11.self_attn.q_proj.kernel", | |
| "model.model.layers.11.self_attn.sinks", | |
| "model.model.layers.11.self_attn.v_proj.bias", | |
| "model.model.layers.11.self_attn.v_proj.kernel", | |
| "model.model.layers.12.input_layernorm.kernel", | |
| "model.model.layers.12.mlp.experts.down_proj.bias", | |
| "model.model.layers.12.mlp.experts.down_proj.kernel", | |
| "model.model.layers.12.mlp.experts.gate_proj.bias", | |
| "model.model.layers.12.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.12.mlp.experts.up_proj.bias", | |
| "model.model.layers.12.mlp.experts.up_proj.kernel", | |
| "model.model.layers.12.mlp.router.bias", | |
| "model.model.layers.12.mlp.router.kernel", | |
| "model.model.layers.12.post_attention_layernorm.kernel", | |
| "model.model.layers.12.self_attn.k_proj.bias", | |
| "model.model.layers.12.self_attn.k_proj.kernel", | |
| "model.model.layers.12.self_attn.o_proj.bias", | |
| "model.model.layers.12.self_attn.o_proj.kernel", | |
| "model.model.layers.12.self_attn.q_proj.bias", | |
| "model.model.layers.12.self_attn.q_proj.kernel", | |
| "model.model.layers.12.self_attn.sinks", | |
| "model.model.layers.12.self_attn.v_proj.bias", | |
| "model.model.layers.12.self_attn.v_proj.kernel", | |
| "model.model.layers.13.input_layernorm.kernel", | |
| "model.model.layers.13.mlp.experts.down_proj.bias", | |
| "model.model.layers.13.mlp.experts.down_proj.kernel", | |
| "model.model.layers.13.mlp.experts.gate_proj.bias", | |
| "model.model.layers.13.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.13.mlp.experts.up_proj.bias", | |
| "model.model.layers.13.mlp.experts.up_proj.kernel", | |
| "model.model.layers.13.mlp.router.bias", | |
| "model.model.layers.13.mlp.router.kernel", | |
| "model.model.layers.13.post_attention_layernorm.kernel", | |
| "model.model.layers.13.self_attn.k_proj.bias", | |
| "model.model.layers.13.self_attn.k_proj.kernel", | |
| "model.model.layers.13.self_attn.o_proj.bias", | |
| "model.model.layers.13.self_attn.o_proj.kernel", | |
| "model.model.layers.13.self_attn.q_proj.bias", | |
| "model.model.layers.13.self_attn.q_proj.kernel", | |
| "model.model.layers.13.self_attn.sinks", | |
| "model.model.layers.13.self_attn.v_proj.bias", | |
| "model.model.layers.13.self_attn.v_proj.kernel", | |
| "model.model.layers.14.input_layernorm.kernel", | |
| "model.model.layers.14.mlp.experts.down_proj.bias", | |
| "model.model.layers.14.mlp.experts.down_proj.kernel", | |
| "model.model.layers.14.mlp.experts.gate_proj.bias", | |
| "model.model.layers.14.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.14.mlp.experts.up_proj.bias", | |
| "model.model.layers.14.mlp.experts.up_proj.kernel", | |
| "model.model.layers.14.mlp.router.bias", | |
| "model.model.layers.14.mlp.router.kernel", | |
| "model.model.layers.14.post_attention_layernorm.kernel", | |
| "model.model.layers.14.self_attn.k_proj.bias", | |
| "model.model.layers.14.self_attn.k_proj.kernel", | |
| "model.model.layers.14.self_attn.o_proj.bias", | |
| "model.model.layers.14.self_attn.o_proj.kernel", | |
| "model.model.layers.14.self_attn.q_proj.bias", | |
| "model.model.layers.14.self_attn.q_proj.kernel", | |
| "model.model.layers.14.self_attn.sinks", | |
| "model.model.layers.14.self_attn.v_proj.bias", | |
| "model.model.layers.14.self_attn.v_proj.kernel", | |
| "model.model.layers.15.input_layernorm.kernel", | |
| "model.model.layers.15.mlp.experts.down_proj.bias", | |
| "model.model.layers.15.mlp.experts.down_proj.kernel", | |
| "model.model.layers.15.mlp.experts.gate_proj.bias", | |
| "model.model.layers.15.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.15.mlp.experts.up_proj.bias", | |
| "model.model.layers.15.mlp.experts.up_proj.kernel", | |
| "model.model.layers.15.mlp.router.bias", | |
| "model.model.layers.15.mlp.router.kernel", | |
| "model.model.layers.15.post_attention_layernorm.kernel", | |
| "model.model.layers.15.self_attn.k_proj.bias", | |
| "model.model.layers.15.self_attn.k_proj.kernel", | |
| "model.model.layers.15.self_attn.o_proj.bias", | |
| "model.model.layers.15.self_attn.o_proj.kernel", | |
| "model.model.layers.15.self_attn.q_proj.bias", | |
| "model.model.layers.15.self_attn.q_proj.kernel", | |
| "model.model.layers.15.self_attn.sinks", | |
| "model.model.layers.15.self_attn.v_proj.bias", | |
| "model.model.layers.15.self_attn.v_proj.kernel", | |
| "model.model.layers.16.input_layernorm.kernel", | |
| "model.model.layers.16.mlp.experts.down_proj.bias", | |
| "model.model.layers.16.mlp.experts.down_proj.kernel", | |
| "model.model.layers.16.mlp.experts.gate_proj.bias", | |
| "model.model.layers.16.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.16.mlp.experts.up_proj.bias", | |
| "model.model.layers.16.mlp.experts.up_proj.kernel", | |
| "model.model.layers.16.mlp.router.bias", | |
| "model.model.layers.16.mlp.router.kernel", | |
| "model.model.layers.16.post_attention_layernorm.kernel", | |
| "model.model.layers.16.self_attn.k_proj.bias", | |
| "model.model.layers.16.self_attn.k_proj.kernel", | |
| "model.model.layers.16.self_attn.o_proj.bias", | |
| "model.model.layers.16.self_attn.o_proj.kernel", | |
| "model.model.layers.16.self_attn.q_proj.bias", | |
| "model.model.layers.16.self_attn.q_proj.kernel", | |
| "model.model.layers.16.self_attn.sinks", | |
| "model.model.layers.16.self_attn.v_proj.bias", | |
| "model.model.layers.16.self_attn.v_proj.kernel", | |
| "model.model.layers.17.input_layernorm.kernel", | |
| "model.model.layers.17.mlp.experts.down_proj.bias", | |
| "model.model.layers.17.mlp.experts.down_proj.kernel", | |
| "model.model.layers.17.mlp.experts.gate_proj.bias", | |
| "model.model.layers.17.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.17.mlp.experts.up_proj.bias", | |
| "model.model.layers.17.mlp.experts.up_proj.kernel", | |
| "model.model.layers.17.mlp.router.bias", | |
| "model.model.layers.17.mlp.router.kernel", | |
| "model.model.layers.17.post_attention_layernorm.kernel", | |
| "model.model.layers.17.self_attn.k_proj.bias", | |
| "model.model.layers.17.self_attn.k_proj.kernel", | |
| "model.model.layers.17.self_attn.o_proj.bias", | |
| "model.model.layers.17.self_attn.o_proj.kernel", | |
| "model.model.layers.17.self_attn.q_proj.bias", | |
| "model.model.layers.17.self_attn.q_proj.kernel", | |
| "model.model.layers.17.self_attn.sinks", | |
| "model.model.layers.17.self_attn.v_proj.bias", | |
| "model.model.layers.17.self_attn.v_proj.kernel", | |
| "model.model.layers.18.input_layernorm.kernel", | |
| "model.model.layers.18.mlp.experts.down_proj.bias", | |
| "model.model.layers.18.mlp.experts.down_proj.kernel", | |
| "model.model.layers.18.mlp.experts.gate_proj.bias", | |
| "model.model.layers.18.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.18.mlp.experts.up_proj.bias", | |
| "model.model.layers.18.mlp.experts.up_proj.kernel", | |
| "model.model.layers.18.mlp.router.bias", | |
| "model.model.layers.18.mlp.router.kernel", | |
| "model.model.layers.18.post_attention_layernorm.kernel", | |
| "model.model.layers.18.self_attn.k_proj.bias", | |
| "model.model.layers.18.self_attn.k_proj.kernel", | |
| "model.model.layers.18.self_attn.o_proj.bias", | |
| "model.model.layers.18.self_attn.o_proj.kernel", | |
| "model.model.layers.18.self_attn.q_proj.bias", | |
| "model.model.layers.18.self_attn.q_proj.kernel", | |
| "model.model.layers.18.self_attn.sinks", | |
| "model.model.layers.18.self_attn.v_proj.bias", | |
| "model.model.layers.18.self_attn.v_proj.kernel", | |
| "model.model.layers.19.input_layernorm.kernel", | |
| "model.model.layers.19.mlp.experts.down_proj.bias", | |
| "model.model.layers.19.mlp.experts.down_proj.kernel", | |
| "model.model.layers.19.mlp.experts.gate_proj.bias", | |
| "model.model.layers.19.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.19.mlp.experts.up_proj.bias", | |
| "model.model.layers.19.mlp.experts.up_proj.kernel", | |
| "model.model.layers.19.mlp.router.bias", | |
| "model.model.layers.19.mlp.router.kernel", | |
| "model.model.layers.19.post_attention_layernorm.kernel", | |
| "model.model.layers.19.self_attn.k_proj.bias", | |
| "model.model.layers.19.self_attn.k_proj.kernel", | |
| "model.model.layers.19.self_attn.o_proj.bias", | |
| "model.model.layers.19.self_attn.o_proj.kernel", | |
| "model.model.layers.19.self_attn.q_proj.bias", | |
| "model.model.layers.19.self_attn.q_proj.kernel", | |
| "model.model.layers.19.self_attn.sinks", | |
| "model.model.layers.19.self_attn.v_proj.bias", | |
| "model.model.layers.19.self_attn.v_proj.kernel", | |
| "model.model.layers.20.input_layernorm.kernel", | |
| "model.model.layers.20.mlp.experts.down_proj.bias", | |
| "model.model.layers.20.mlp.experts.down_proj.kernel", | |
| "model.model.layers.20.mlp.experts.gate_proj.bias", | |
| "model.model.layers.20.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.20.mlp.experts.up_proj.bias", | |
| "model.model.layers.20.mlp.experts.up_proj.kernel", | |
| "model.model.layers.20.mlp.router.bias", | |
| "model.model.layers.20.mlp.router.kernel", | |
| "model.model.layers.20.post_attention_layernorm.kernel", | |
| "model.model.layers.20.self_attn.k_proj.bias", | |
| "model.model.layers.20.self_attn.k_proj.kernel", | |
| "model.model.layers.20.self_attn.o_proj.bias", | |
| "model.model.layers.20.self_attn.o_proj.kernel", | |
| "model.model.layers.20.self_attn.q_proj.bias", | |
| "model.model.layers.20.self_attn.q_proj.kernel", | |
| "model.model.layers.20.self_attn.sinks", | |
| "model.model.layers.20.self_attn.v_proj.bias", | |
| "model.model.layers.20.self_attn.v_proj.kernel", | |
| "model.model.layers.21.input_layernorm.kernel", | |
| "model.model.layers.21.mlp.experts.down_proj.bias", | |
| "model.model.layers.21.mlp.experts.down_proj.kernel", | |
| "model.model.layers.21.mlp.experts.gate_proj.bias", | |
| "model.model.layers.21.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.21.mlp.experts.up_proj.bias", | |
| "model.model.layers.21.mlp.experts.up_proj.kernel", | |
| "model.model.layers.21.mlp.router.bias", | |
| "model.model.layers.21.mlp.router.kernel", | |
| "model.model.layers.21.post_attention_layernorm.kernel", | |
| "model.model.layers.21.self_attn.k_proj.bias", | |
| "model.model.layers.21.self_attn.k_proj.kernel", | |
| "model.model.layers.21.self_attn.o_proj.bias", | |
| "model.model.layers.21.self_attn.o_proj.kernel", | |
| "model.model.layers.21.self_attn.q_proj.bias", | |
| "model.model.layers.21.self_attn.q_proj.kernel", | |
| "model.model.layers.21.self_attn.sinks", | |
| "model.model.layers.21.self_attn.v_proj.bias", | |
| "model.model.layers.21.self_attn.v_proj.kernel", | |
| "model.model.layers.22.input_layernorm.kernel", | |
| "model.model.layers.22.mlp.experts.down_proj.bias", | |
| "model.model.layers.22.mlp.experts.down_proj.kernel", | |
| "model.model.layers.22.mlp.experts.gate_proj.bias", | |
| "model.model.layers.22.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.22.mlp.experts.up_proj.bias", | |
| "model.model.layers.22.mlp.experts.up_proj.kernel", | |
| "model.model.layers.22.mlp.router.bias", | |
| "model.model.layers.22.mlp.router.kernel", | |
| "model.model.layers.22.post_attention_layernorm.kernel", | |
| "model.model.layers.22.self_attn.k_proj.bias", | |
| "model.model.layers.22.self_attn.k_proj.kernel", | |
| "model.model.layers.22.self_attn.o_proj.bias", | |
| "model.model.layers.22.self_attn.o_proj.kernel", | |
| "model.model.layers.22.self_attn.q_proj.bias", | |
| "model.model.layers.22.self_attn.q_proj.kernel", | |
| "model.model.layers.22.self_attn.sinks", | |
| "model.model.layers.22.self_attn.v_proj.bias", | |
| "model.model.layers.22.self_attn.v_proj.kernel", | |
| "model.model.layers.23.input_layernorm.kernel", | |
| "model.model.layers.23.mlp.experts.down_proj.bias", | |
| "model.model.layers.23.mlp.experts.down_proj.kernel", | |
| "model.model.layers.23.mlp.experts.gate_proj.bias", | |
| "model.model.layers.23.mlp.experts.gate_proj.kernel", | |
| "model.model.layers.23.mlp.experts.up_proj.bias", | |
| "model.model.layers.23.mlp.experts.up_proj.kernel", | |
| "model.model.layers.23.mlp.router.bias", | |
| "model.model.layers.23.mlp.router.kernel", | |
| "model.model.layers.23.post_attention_layernorm.kernel", | |
| "model.model.layers.23.self_attn.k_proj.bias", | |
| "model.model.layers.23.self_attn.k_proj.kernel", | |
| "model.model.layers.23.self_attn.o_proj.bias", | |
| "model.model.layers.23.self_attn.o_proj.kernel", | |
| "model.model.layers.23.self_attn.q_proj.bias", | |
| "model.model.layers.23.self_attn.q_proj.kernel", | |
| "model.model.layers.23.self_attn.sinks", | |
| "model.model.layers.23.self_attn.v_proj.bias", | |
| "model.model.layers.23.self_attn.v_proj.kernel", | |
| "model.model.norm.kernel" | |
| ], | |
| "array_relpaths": [ | |
| "model/lm_head/kernel", | |
| "model/model/embed_tokens/embedding", | |
| "model/model/layers/0/input_layernorm/kernel", | |
| "model/model/layers/0/mlp/experts/down_proj/bias", | |
| "model/model/layers/0/mlp/experts/down_proj/kernel", | |
| "model/model/layers/0/mlp/experts/gate_proj/bias", | |
| "model/model/layers/0/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/0/mlp/experts/up_proj/bias", | |
| "model/model/layers/0/mlp/experts/up_proj/kernel", | |
| "model/model/layers/0/mlp/router/bias", | |
| "model/model/layers/0/mlp/router/kernel", | |
| "model/model/layers/0/post_attention_layernorm/kernel", | |
| "model/model/layers/0/self_attn/k_proj/bias", | |
| "model/model/layers/0/self_attn/k_proj/kernel", | |
| "model/model/layers/0/self_attn/o_proj/bias", | |
| "model/model/layers/0/self_attn/o_proj/kernel", | |
| "model/model/layers/0/self_attn/q_proj/bias", | |
| "model/model/layers/0/self_attn/q_proj/kernel", | |
| "model/model/layers/0/self_attn/sinks", | |
| "model/model/layers/0/self_attn/v_proj/bias", | |
| "model/model/layers/0/self_attn/v_proj/kernel", | |
| "model/model/layers/1/input_layernorm/kernel", | |
| "model/model/layers/1/mlp/experts/down_proj/bias", | |
| "model/model/layers/1/mlp/experts/down_proj/kernel", | |
| "model/model/layers/1/mlp/experts/gate_proj/bias", | |
| "model/model/layers/1/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/1/mlp/experts/up_proj/bias", | |
| "model/model/layers/1/mlp/experts/up_proj/kernel", | |
| "model/model/layers/1/mlp/router/bias", | |
| "model/model/layers/1/mlp/router/kernel", | |
| "model/model/layers/1/post_attention_layernorm/kernel", | |
| "model/model/layers/1/self_attn/k_proj/bias", | |
| "model/model/layers/1/self_attn/k_proj/kernel", | |
| "model/model/layers/1/self_attn/o_proj/bias", | |
| "model/model/layers/1/self_attn/o_proj/kernel", | |
| "model/model/layers/1/self_attn/q_proj/bias", | |
| "model/model/layers/1/self_attn/q_proj/kernel", | |
| "model/model/layers/1/self_attn/sinks", | |
| "model/model/layers/1/self_attn/v_proj/bias", | |
| "model/model/layers/1/self_attn/v_proj/kernel", | |
| "model/model/layers/2/input_layernorm/kernel", | |
| "model/model/layers/2/mlp/experts/down_proj/bias", | |
| "model/model/layers/2/mlp/experts/down_proj/kernel", | |
| "model/model/layers/2/mlp/experts/gate_proj/bias", | |
| "model/model/layers/2/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/2/mlp/experts/up_proj/bias", | |
| "model/model/layers/2/mlp/experts/up_proj/kernel", | |
| "model/model/layers/2/mlp/router/bias", | |
| "model/model/layers/2/mlp/router/kernel", | |
| "model/model/layers/2/post_attention_layernorm/kernel", | |
| "model/model/layers/2/self_attn/k_proj/bias", | |
| "model/model/layers/2/self_attn/k_proj/kernel", | |
| "model/model/layers/2/self_attn/o_proj/bias", | |
| "model/model/layers/2/self_attn/o_proj/kernel", | |
| "model/model/layers/2/self_attn/q_proj/bias", | |
| "model/model/layers/2/self_attn/q_proj/kernel", | |
| "model/model/layers/2/self_attn/sinks", | |
| "model/model/layers/2/self_attn/v_proj/bias", | |
| "model/model/layers/2/self_attn/v_proj/kernel", | |
| "model/model/layers/3/input_layernorm/kernel", | |
| "model/model/layers/3/mlp/experts/down_proj/bias", | |
| "model/model/layers/3/mlp/experts/down_proj/kernel", | |
| "model/model/layers/3/mlp/experts/gate_proj/bias", | |
| "model/model/layers/3/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/3/mlp/experts/up_proj/bias", | |
| "model/model/layers/3/mlp/experts/up_proj/kernel", | |
| "model/model/layers/3/mlp/router/bias", | |
| "model/model/layers/3/mlp/router/kernel", | |
| "model/model/layers/3/post_attention_layernorm/kernel", | |
| "model/model/layers/3/self_attn/k_proj/bias", | |
| "model/model/layers/3/self_attn/k_proj/kernel", | |
| "model/model/layers/3/self_attn/o_proj/bias", | |
| "model/model/layers/3/self_attn/o_proj/kernel", | |
| "model/model/layers/3/self_attn/q_proj/bias", | |
| "model/model/layers/3/self_attn/q_proj/kernel", | |
| "model/model/layers/3/self_attn/sinks", | |
| "model/model/layers/3/self_attn/v_proj/bias", | |
| "model/model/layers/3/self_attn/v_proj/kernel", | |
| "model/model/layers/4/input_layernorm/kernel", | |
| "model/model/layers/4/mlp/experts/down_proj/bias", | |
| "model/model/layers/4/mlp/experts/down_proj/kernel", | |
| "model/model/layers/4/mlp/experts/gate_proj/bias", | |
| "model/model/layers/4/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/4/mlp/experts/up_proj/bias", | |
| "model/model/layers/4/mlp/experts/up_proj/kernel", | |
| "model/model/layers/4/mlp/router/bias", | |
| "model/model/layers/4/mlp/router/kernel", | |
| "model/model/layers/4/post_attention_layernorm/kernel", | |
| "model/model/layers/4/self_attn/k_proj/bias", | |
| "model/model/layers/4/self_attn/k_proj/kernel", | |
| "model/model/layers/4/self_attn/o_proj/bias", | |
| "model/model/layers/4/self_attn/o_proj/kernel", | |
| "model/model/layers/4/self_attn/q_proj/bias", | |
| "model/model/layers/4/self_attn/q_proj/kernel", | |
| "model/model/layers/4/self_attn/sinks", | |
| "model/model/layers/4/self_attn/v_proj/bias", | |
| "model/model/layers/4/self_attn/v_proj/kernel", | |
| "model/model/layers/5/input_layernorm/kernel", | |
| "model/model/layers/5/mlp/experts/down_proj/bias", | |
| "model/model/layers/5/mlp/experts/down_proj/kernel", | |
| "model/model/layers/5/mlp/experts/gate_proj/bias", | |
| "model/model/layers/5/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/5/mlp/experts/up_proj/bias", | |
| "model/model/layers/5/mlp/experts/up_proj/kernel", | |
| "model/model/layers/5/mlp/router/bias", | |
| "model/model/layers/5/mlp/router/kernel", | |
| "model/model/layers/5/post_attention_layernorm/kernel", | |
| "model/model/layers/5/self_attn/k_proj/bias", | |
| "model/model/layers/5/self_attn/k_proj/kernel", | |
| "model/model/layers/5/self_attn/o_proj/bias", | |
| "model/model/layers/5/self_attn/o_proj/kernel", | |
| "model/model/layers/5/self_attn/q_proj/bias", | |
| "model/model/layers/5/self_attn/q_proj/kernel", | |
| "model/model/layers/5/self_attn/sinks", | |
| "model/model/layers/5/self_attn/v_proj/bias", | |
| "model/model/layers/5/self_attn/v_proj/kernel", | |
| "model/model/layers/6/input_layernorm/kernel", | |
| "model/model/layers/6/mlp/experts/down_proj/bias", | |
| "model/model/layers/6/mlp/experts/down_proj/kernel", | |
| "model/model/layers/6/mlp/experts/gate_proj/bias", | |
| "model/model/layers/6/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/6/mlp/experts/up_proj/bias", | |
| "model/model/layers/6/mlp/experts/up_proj/kernel", | |
| "model/model/layers/6/mlp/router/bias", | |
| "model/model/layers/6/mlp/router/kernel", | |
| "model/model/layers/6/post_attention_layernorm/kernel", | |
| "model/model/layers/6/self_attn/k_proj/bias", | |
| "model/model/layers/6/self_attn/k_proj/kernel", | |
| "model/model/layers/6/self_attn/o_proj/bias", | |
| "model/model/layers/6/self_attn/o_proj/kernel", | |
| "model/model/layers/6/self_attn/q_proj/bias", | |
| "model/model/layers/6/self_attn/q_proj/kernel", | |
| "model/model/layers/6/self_attn/sinks", | |
| "model/model/layers/6/self_attn/v_proj/bias", | |
| "model/model/layers/6/self_attn/v_proj/kernel", | |
| "model/model/layers/7/input_layernorm/kernel", | |
| "model/model/layers/7/mlp/experts/down_proj/bias", | |
| "model/model/layers/7/mlp/experts/down_proj/kernel", | |
| "model/model/layers/7/mlp/experts/gate_proj/bias", | |
| "model/model/layers/7/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/7/mlp/experts/up_proj/bias", | |
| "model/model/layers/7/mlp/experts/up_proj/kernel", | |
| "model/model/layers/7/mlp/router/bias", | |
| "model/model/layers/7/mlp/router/kernel", | |
| "model/model/layers/7/post_attention_layernorm/kernel", | |
| "model/model/layers/7/self_attn/k_proj/bias", | |
| "model/model/layers/7/self_attn/k_proj/kernel", | |
| "model/model/layers/7/self_attn/o_proj/bias", | |
| "model/model/layers/7/self_attn/o_proj/kernel", | |
| "model/model/layers/7/self_attn/q_proj/bias", | |
| "model/model/layers/7/self_attn/q_proj/kernel", | |
| "model/model/layers/7/self_attn/sinks", | |
| "model/model/layers/7/self_attn/v_proj/bias", | |
| "model/model/layers/7/self_attn/v_proj/kernel", | |
| "model/model/layers/8/input_layernorm/kernel", | |
| "model/model/layers/8/mlp/experts/down_proj/bias", | |
| "model/model/layers/8/mlp/experts/down_proj/kernel", | |
| "model/model/layers/8/mlp/experts/gate_proj/bias", | |
| "model/model/layers/8/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/8/mlp/experts/up_proj/bias", | |
| "model/model/layers/8/mlp/experts/up_proj/kernel", | |
| "model/model/layers/8/mlp/router/bias", | |
| "model/model/layers/8/mlp/router/kernel", | |
| "model/model/layers/8/post_attention_layernorm/kernel", | |
| "model/model/layers/8/self_attn/k_proj/bias", | |
| "model/model/layers/8/self_attn/k_proj/kernel", | |
| "model/model/layers/8/self_attn/o_proj/bias", | |
| "model/model/layers/8/self_attn/o_proj/kernel", | |
| "model/model/layers/8/self_attn/q_proj/bias", | |
| "model/model/layers/8/self_attn/q_proj/kernel", | |
| "model/model/layers/8/self_attn/sinks", | |
| "model/model/layers/8/self_attn/v_proj/bias", | |
| "model/model/layers/8/self_attn/v_proj/kernel", | |
| "model/model/layers/9/input_layernorm/kernel", | |
| "model/model/layers/9/mlp/experts/down_proj/bias", | |
| "model/model/layers/9/mlp/experts/down_proj/kernel", | |
| "model/model/layers/9/mlp/experts/gate_proj/bias", | |
| "model/model/layers/9/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/9/mlp/experts/up_proj/bias", | |
| "model/model/layers/9/mlp/experts/up_proj/kernel", | |
| "model/model/layers/9/mlp/router/bias", | |
| "model/model/layers/9/mlp/router/kernel", | |
| "model/model/layers/9/post_attention_layernorm/kernel", | |
| "model/model/layers/9/self_attn/k_proj/bias", | |
| "model/model/layers/9/self_attn/k_proj/kernel", | |
| "model/model/layers/9/self_attn/o_proj/bias", | |
| "model/model/layers/9/self_attn/o_proj/kernel", | |
| "model/model/layers/9/self_attn/q_proj/bias", | |
| "model/model/layers/9/self_attn/q_proj/kernel", | |
| "model/model/layers/9/self_attn/sinks", | |
| "model/model/layers/9/self_attn/v_proj/bias", | |
| "model/model/layers/9/self_attn/v_proj/kernel", | |
| "model/model/layers/10/input_layernorm/kernel", | |
| "model/model/layers/10/mlp/experts/down_proj/bias", | |
| "model/model/layers/10/mlp/experts/down_proj/kernel", | |
| "model/model/layers/10/mlp/experts/gate_proj/bias", | |
| "model/model/layers/10/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/10/mlp/experts/up_proj/bias", | |
| "model/model/layers/10/mlp/experts/up_proj/kernel", | |
| "model/model/layers/10/mlp/router/bias", | |
| "model/model/layers/10/mlp/router/kernel", | |
| "model/model/layers/10/post_attention_layernorm/kernel", | |
| "model/model/layers/10/self_attn/k_proj/bias", | |
| "model/model/layers/10/self_attn/k_proj/kernel", | |
| "model/model/layers/10/self_attn/o_proj/bias", | |
| "model/model/layers/10/self_attn/o_proj/kernel", | |
| "model/model/layers/10/self_attn/q_proj/bias", | |
| "model/model/layers/10/self_attn/q_proj/kernel", | |
| "model/model/layers/10/self_attn/sinks", | |
| "model/model/layers/10/self_attn/v_proj/bias", | |
| "model/model/layers/10/self_attn/v_proj/kernel", | |
| "model/model/layers/11/input_layernorm/kernel", | |
| "model/model/layers/11/mlp/experts/down_proj/bias", | |
| "model/model/layers/11/mlp/experts/down_proj/kernel", | |
| "model/model/layers/11/mlp/experts/gate_proj/bias", | |
| "model/model/layers/11/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/11/mlp/experts/up_proj/bias", | |
| "model/model/layers/11/mlp/experts/up_proj/kernel", | |
| "model/model/layers/11/mlp/router/bias", | |
| "model/model/layers/11/mlp/router/kernel", | |
| "model/model/layers/11/post_attention_layernorm/kernel", | |
| "model/model/layers/11/self_attn/k_proj/bias", | |
| "model/model/layers/11/self_attn/k_proj/kernel", | |
| "model/model/layers/11/self_attn/o_proj/bias", | |
| "model/model/layers/11/self_attn/o_proj/kernel", | |
| "model/model/layers/11/self_attn/q_proj/bias", | |
| "model/model/layers/11/self_attn/q_proj/kernel", | |
| "model/model/layers/11/self_attn/sinks", | |
| "model/model/layers/11/self_attn/v_proj/bias", | |
| "model/model/layers/11/self_attn/v_proj/kernel", | |
| "model/model/layers/12/input_layernorm/kernel", | |
| "model/model/layers/12/mlp/experts/down_proj/bias", | |
| "model/model/layers/12/mlp/experts/down_proj/kernel", | |
| "model/model/layers/12/mlp/experts/gate_proj/bias", | |
| "model/model/layers/12/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/12/mlp/experts/up_proj/bias", | |
| "model/model/layers/12/mlp/experts/up_proj/kernel", | |
| "model/model/layers/12/mlp/router/bias", | |
| "model/model/layers/12/mlp/router/kernel", | |
| "model/model/layers/12/post_attention_layernorm/kernel", | |
| "model/model/layers/12/self_attn/k_proj/bias", | |
| "model/model/layers/12/self_attn/k_proj/kernel", | |
| "model/model/layers/12/self_attn/o_proj/bias", | |
| "model/model/layers/12/self_attn/o_proj/kernel", | |
| "model/model/layers/12/self_attn/q_proj/bias", | |
| "model/model/layers/12/self_attn/q_proj/kernel", | |
| "model/model/layers/12/self_attn/sinks", | |
| "model/model/layers/12/self_attn/v_proj/bias", | |
| "model/model/layers/12/self_attn/v_proj/kernel", | |
| "model/model/layers/13/input_layernorm/kernel", | |
| "model/model/layers/13/mlp/experts/down_proj/bias", | |
| "model/model/layers/13/mlp/experts/down_proj/kernel", | |
| "model/model/layers/13/mlp/experts/gate_proj/bias", | |
| "model/model/layers/13/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/13/mlp/experts/up_proj/bias", | |
| "model/model/layers/13/mlp/experts/up_proj/kernel", | |
| "model/model/layers/13/mlp/router/bias", | |
| "model/model/layers/13/mlp/router/kernel", | |
| "model/model/layers/13/post_attention_layernorm/kernel", | |
| "model/model/layers/13/self_attn/k_proj/bias", | |
| "model/model/layers/13/self_attn/k_proj/kernel", | |
| "model/model/layers/13/self_attn/o_proj/bias", | |
| "model/model/layers/13/self_attn/o_proj/kernel", | |
| "model/model/layers/13/self_attn/q_proj/bias", | |
| "model/model/layers/13/self_attn/q_proj/kernel", | |
| "model/model/layers/13/self_attn/sinks", | |
| "model/model/layers/13/self_attn/v_proj/bias", | |
| "model/model/layers/13/self_attn/v_proj/kernel", | |
| "model/model/layers/14/input_layernorm/kernel", | |
| "model/model/layers/14/mlp/experts/down_proj/bias", | |
| "model/model/layers/14/mlp/experts/down_proj/kernel", | |
| "model/model/layers/14/mlp/experts/gate_proj/bias", | |
| "model/model/layers/14/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/14/mlp/experts/up_proj/bias", | |
| "model/model/layers/14/mlp/experts/up_proj/kernel", | |
| "model/model/layers/14/mlp/router/bias", | |
| "model/model/layers/14/mlp/router/kernel", | |
| "model/model/layers/14/post_attention_layernorm/kernel", | |
| "model/model/layers/14/self_attn/k_proj/bias", | |
| "model/model/layers/14/self_attn/k_proj/kernel", | |
| "model/model/layers/14/self_attn/o_proj/bias", | |
| "model/model/layers/14/self_attn/o_proj/kernel", | |
| "model/model/layers/14/self_attn/q_proj/bias", | |
| "model/model/layers/14/self_attn/q_proj/kernel", | |
| "model/model/layers/14/self_attn/sinks", | |
| "model/model/layers/14/self_attn/v_proj/bias", | |
| "model/model/layers/14/self_attn/v_proj/kernel", | |
| "model/model/layers/15/input_layernorm/kernel", | |
| "model/model/layers/15/mlp/experts/down_proj/bias", | |
| "model/model/layers/15/mlp/experts/down_proj/kernel", | |
| "model/model/layers/15/mlp/experts/gate_proj/bias", | |
| "model/model/layers/15/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/15/mlp/experts/up_proj/bias", | |
| "model/model/layers/15/mlp/experts/up_proj/kernel", | |
| "model/model/layers/15/mlp/router/bias", | |
| "model/model/layers/15/mlp/router/kernel", | |
| "model/model/layers/15/post_attention_layernorm/kernel", | |
| "model/model/layers/15/self_attn/k_proj/bias", | |
| "model/model/layers/15/self_attn/k_proj/kernel", | |
| "model/model/layers/15/self_attn/o_proj/bias", | |
| "model/model/layers/15/self_attn/o_proj/kernel", | |
| "model/model/layers/15/self_attn/q_proj/bias", | |
| "model/model/layers/15/self_attn/q_proj/kernel", | |
| "model/model/layers/15/self_attn/sinks", | |
| "model/model/layers/15/self_attn/v_proj/bias", | |
| "model/model/layers/15/self_attn/v_proj/kernel", | |
| "model/model/layers/16/input_layernorm/kernel", | |
| "model/model/layers/16/mlp/experts/down_proj/bias", | |
| "model/model/layers/16/mlp/experts/down_proj/kernel", | |
| "model/model/layers/16/mlp/experts/gate_proj/bias", | |
| "model/model/layers/16/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/16/mlp/experts/up_proj/bias", | |
| "model/model/layers/16/mlp/experts/up_proj/kernel", | |
| "model/model/layers/16/mlp/router/bias", | |
| "model/model/layers/16/mlp/router/kernel", | |
| "model/model/layers/16/post_attention_layernorm/kernel", | |
| "model/model/layers/16/self_attn/k_proj/bias", | |
| "model/model/layers/16/self_attn/k_proj/kernel", | |
| "model/model/layers/16/self_attn/o_proj/bias", | |
| "model/model/layers/16/self_attn/o_proj/kernel", | |
| "model/model/layers/16/self_attn/q_proj/bias", | |
| "model/model/layers/16/self_attn/q_proj/kernel", | |
| "model/model/layers/16/self_attn/sinks", | |
| "model/model/layers/16/self_attn/v_proj/bias", | |
| "model/model/layers/16/self_attn/v_proj/kernel", | |
| "model/model/layers/17/input_layernorm/kernel", | |
| "model/model/layers/17/mlp/experts/down_proj/bias", | |
| "model/model/layers/17/mlp/experts/down_proj/kernel", | |
| "model/model/layers/17/mlp/experts/gate_proj/bias", | |
| "model/model/layers/17/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/17/mlp/experts/up_proj/bias", | |
| "model/model/layers/17/mlp/experts/up_proj/kernel", | |
| "model/model/layers/17/mlp/router/bias", | |
| "model/model/layers/17/mlp/router/kernel", | |
| "model/model/layers/17/post_attention_layernorm/kernel", | |
| "model/model/layers/17/self_attn/k_proj/bias", | |
| "model/model/layers/17/self_attn/k_proj/kernel", | |
| "model/model/layers/17/self_attn/o_proj/bias", | |
| "model/model/layers/17/self_attn/o_proj/kernel", | |
| "model/model/layers/17/self_attn/q_proj/bias", | |
| "model/model/layers/17/self_attn/q_proj/kernel", | |
| "model/model/layers/17/self_attn/sinks", | |
| "model/model/layers/17/self_attn/v_proj/bias", | |
| "model/model/layers/17/self_attn/v_proj/kernel", | |
| "model/model/layers/18/input_layernorm/kernel", | |
| "model/model/layers/18/mlp/experts/down_proj/bias", | |
| "model/model/layers/18/mlp/experts/down_proj/kernel", | |
| "model/model/layers/18/mlp/experts/gate_proj/bias", | |
| "model/model/layers/18/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/18/mlp/experts/up_proj/bias", | |
| "model/model/layers/18/mlp/experts/up_proj/kernel", | |
| "model/model/layers/18/mlp/router/bias", | |
| "model/model/layers/18/mlp/router/kernel", | |
| "model/model/layers/18/post_attention_layernorm/kernel", | |
| "model/model/layers/18/self_attn/k_proj/bias", | |
| "model/model/layers/18/self_attn/k_proj/kernel", | |
| "model/model/layers/18/self_attn/o_proj/bias", | |
| "model/model/layers/18/self_attn/o_proj/kernel", | |
| "model/model/layers/18/self_attn/q_proj/bias", | |
| "model/model/layers/18/self_attn/q_proj/kernel", | |
| "model/model/layers/18/self_attn/sinks", | |
| "model/model/layers/18/self_attn/v_proj/bias", | |
| "model/model/layers/18/self_attn/v_proj/kernel", | |
| "model/model/layers/19/input_layernorm/kernel", | |
| "model/model/layers/19/mlp/experts/down_proj/bias", | |
| "model/model/layers/19/mlp/experts/down_proj/kernel", | |
| "model/model/layers/19/mlp/experts/gate_proj/bias", | |
| "model/model/layers/19/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/19/mlp/experts/up_proj/bias", | |
| "model/model/layers/19/mlp/experts/up_proj/kernel", | |
| "model/model/layers/19/mlp/router/bias", | |
| "model/model/layers/19/mlp/router/kernel", | |
| "model/model/layers/19/post_attention_layernorm/kernel", | |
| "model/model/layers/19/self_attn/k_proj/bias", | |
| "model/model/layers/19/self_attn/k_proj/kernel", | |
| "model/model/layers/19/self_attn/o_proj/bias", | |
| "model/model/layers/19/self_attn/o_proj/kernel", | |
| "model/model/layers/19/self_attn/q_proj/bias", | |
| "model/model/layers/19/self_attn/q_proj/kernel", | |
| "model/model/layers/19/self_attn/sinks", | |
| "model/model/layers/19/self_attn/v_proj/bias", | |
| "model/model/layers/19/self_attn/v_proj/kernel", | |
| "model/model/layers/20/input_layernorm/kernel", | |
| "model/model/layers/20/mlp/experts/down_proj/bias", | |
| "model/model/layers/20/mlp/experts/down_proj/kernel", | |
| "model/model/layers/20/mlp/experts/gate_proj/bias", | |
| "model/model/layers/20/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/20/mlp/experts/up_proj/bias", | |
| "model/model/layers/20/mlp/experts/up_proj/kernel", | |
| "model/model/layers/20/mlp/router/bias", | |
| "model/model/layers/20/mlp/router/kernel", | |
| "model/model/layers/20/post_attention_layernorm/kernel", | |
| "model/model/layers/20/self_attn/k_proj/bias", | |
| "model/model/layers/20/self_attn/k_proj/kernel", | |
| "model/model/layers/20/self_attn/o_proj/bias", | |
| "model/model/layers/20/self_attn/o_proj/kernel", | |
| "model/model/layers/20/self_attn/q_proj/bias", | |
| "model/model/layers/20/self_attn/q_proj/kernel", | |
| "model/model/layers/20/self_attn/sinks", | |
| "model/model/layers/20/self_attn/v_proj/bias", | |
| "model/model/layers/20/self_attn/v_proj/kernel", | |
| "model/model/layers/21/input_layernorm/kernel", | |
| "model/model/layers/21/mlp/experts/down_proj/bias", | |
| "model/model/layers/21/mlp/experts/down_proj/kernel", | |
| "model/model/layers/21/mlp/experts/gate_proj/bias", | |
| "model/model/layers/21/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/21/mlp/experts/up_proj/bias", | |
| "model/model/layers/21/mlp/experts/up_proj/kernel", | |
| "model/model/layers/21/mlp/router/bias", | |
| "model/model/layers/21/mlp/router/kernel", | |
| "model/model/layers/21/post_attention_layernorm/kernel", | |
| "model/model/layers/21/self_attn/k_proj/bias", | |
| "model/model/layers/21/self_attn/k_proj/kernel", | |
| "model/model/layers/21/self_attn/o_proj/bias", | |
| "model/model/layers/21/self_attn/o_proj/kernel", | |
| "model/model/layers/21/self_attn/q_proj/bias", | |
| "model/model/layers/21/self_attn/q_proj/kernel", | |
| "model/model/layers/21/self_attn/sinks", | |
| "model/model/layers/21/self_attn/v_proj/bias", | |
| "model/model/layers/21/self_attn/v_proj/kernel", | |
| "model/model/layers/22/input_layernorm/kernel", | |
| "model/model/layers/22/mlp/experts/down_proj/bias", | |
| "model/model/layers/22/mlp/experts/down_proj/kernel", | |
| "model/model/layers/22/mlp/experts/gate_proj/bias", | |
| "model/model/layers/22/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/22/mlp/experts/up_proj/bias", | |
| "model/model/layers/22/mlp/experts/up_proj/kernel", | |
| "model/model/layers/22/mlp/router/bias", | |
| "model/model/layers/22/mlp/router/kernel", | |
| "model/model/layers/22/post_attention_layernorm/kernel", | |
| "model/model/layers/22/self_attn/k_proj/bias", | |
| "model/model/layers/22/self_attn/k_proj/kernel", | |
| "model/model/layers/22/self_attn/o_proj/bias", | |
| "model/model/layers/22/self_attn/o_proj/kernel", | |
| "model/model/layers/22/self_attn/q_proj/bias", | |
| "model/model/layers/22/self_attn/q_proj/kernel", | |
| "model/model/layers/22/self_attn/sinks", | |
| "model/model/layers/22/self_attn/v_proj/bias", | |
| "model/model/layers/22/self_attn/v_proj/kernel", | |
| "model/model/layers/23/input_layernorm/kernel", | |
| "model/model/layers/23/mlp/experts/down_proj/bias", | |
| "model/model/layers/23/mlp/experts/down_proj/kernel", | |
| "model/model/layers/23/mlp/experts/gate_proj/bias", | |
| "model/model/layers/23/mlp/experts/gate_proj/kernel", | |
| "model/model/layers/23/mlp/experts/up_proj/bias", | |
| "model/model/layers/23/mlp/experts/up_proj/kernel", | |
| "model/model/layers/23/mlp/router/bias", | |
| "model/model/layers/23/mlp/router/kernel", | |
| "model/model/layers/23/post_attention_layernorm/kernel", | |
| "model/model/layers/23/self_attn/k_proj/bias", | |
| "model/model/layers/23/self_attn/k_proj/kernel", | |
| "model/model/layers/23/self_attn/o_proj/bias", | |
| "model/model/layers/23/self_attn/o_proj/kernel", | |
| "model/model/layers/23/self_attn/q_proj/bias", | |
| "model/model/layers/23/self_attn/q_proj/kernel", | |
| "model/model/layers/23/self_attn/sinks", | |
| "model/model/layers/23/self_attn/v_proj/bias", | |
| "model/model/layers/23/self_attn/v_proj/kernel", | |
| "model/model/norm/kernel" | |
| ], | |
| "nonarray_payload": {}, | |
| "safetensors_file": null, | |
| "extras": {} | |
| } |