{ "metadata": { "total_size": 774832176.0 }, "weight_map": { "lm_head.weight": "pytorch_model-00008-of-00008.bin", "transformer.h.0.attn.bias": "pytorch_model-00003-of-00008.bin", "transformer.h.0.attn.c_attn.bias": "pytorch_model-00003-of-00008.bin", "transformer.h.0.attn.c_attn.weight": "pytorch_model-00003-of-00008.bin", "transformer.h.0.attn.c_proj.bias": "pytorch_model-00003-of-00008.bin", "transformer.h.0.attn.c_proj.weight": "pytorch_model-00003-of-00008.bin", "transformer.h.0.attn.masked_bias": "pytorch_model-00003-of-00008.bin", "transformer.h.0.ln_1.bias": "pytorch_model-00003-of-00008.bin", "transformer.h.0.ln_1.weight": "pytorch_model-00003-of-00008.bin", "transformer.h.0.ln_2.bias": "pytorch_model-00003-of-00008.bin", "transformer.h.0.ln_2.weight": "pytorch_model-00003-of-00008.bin", "transformer.h.0.mlp.c_fc.bias": "pytorch_model-00003-of-00008.bin", "transformer.h.0.mlp.c_fc.weight": "pytorch_model-00003-of-00008.bin", "transformer.h.0.mlp.c_proj.bias": "pytorch_model-00003-of-00008.bin", "transformer.h.0.mlp.c_proj.weight": "pytorch_model-00003-of-00008.bin", "transformer.h.1.attn.bias": "pytorch_model-00003-of-00008.bin", "transformer.h.1.attn.c_attn.bias": "pytorch_model-00003-of-00008.bin", "transformer.h.1.attn.c_attn.weight": "pytorch_model-00003-of-00008.bin", "transformer.h.1.attn.c_proj.bias": "pytorch_model-00003-of-00008.bin", "transformer.h.1.attn.c_proj.weight": "pytorch_model-00003-of-00008.bin", "transformer.h.1.attn.masked_bias": "pytorch_model-00003-of-00008.bin", "transformer.h.1.ln_1.bias": "pytorch_model-00003-of-00008.bin", "transformer.h.1.ln_1.weight": "pytorch_model-00003-of-00008.bin", "transformer.h.1.ln_2.bias": "pytorch_model-00003-of-00008.bin", "transformer.h.1.ln_2.weight": "pytorch_model-00003-of-00008.bin", "transformer.h.1.mlp.c_fc.bias": "pytorch_model-00003-of-00008.bin", "transformer.h.1.mlp.c_fc.weight": "pytorch_model-00003-of-00008.bin", "transformer.h.1.mlp.c_proj.bias": "pytorch_model-00003-of-00008.bin", "transformer.h.1.mlp.c_proj.weight": "pytorch_model-00003-of-00008.bin", "transformer.h.10.attn.bias": "pytorch_model-00007-of-00008.bin", "transformer.h.10.attn.c_attn.bias": "pytorch_model-00007-of-00008.bin", "transformer.h.10.attn.c_attn.weight": "pytorch_model-00007-of-00008.bin", "transformer.h.10.attn.c_proj.bias": "pytorch_model-00007-of-00008.bin", "transformer.h.10.attn.c_proj.weight": "pytorch_model-00007-of-00008.bin", "transformer.h.10.attn.masked_bias": "pytorch_model-00007-of-00008.bin", "transformer.h.10.ln_1.bias": "pytorch_model-00007-of-00008.bin", "transformer.h.10.ln_1.weight": "pytorch_model-00007-of-00008.bin", "transformer.h.10.ln_2.bias": "pytorch_model-00007-of-00008.bin", "transformer.h.10.ln_2.weight": "pytorch_model-00007-of-00008.bin", "transformer.h.10.mlp.c_fc.bias": "pytorch_model-00007-of-00008.bin", "transformer.h.10.mlp.c_fc.weight": "pytorch_model-00007-of-00008.bin", "transformer.h.10.mlp.c_proj.bias": "pytorch_model-00007-of-00008.bin", "transformer.h.10.mlp.c_proj.weight": "pytorch_model-00007-of-00008.bin", "transformer.h.11.attn.bias": "pytorch_model-00007-of-00008.bin", "transformer.h.11.attn.c_attn.bias": "pytorch_model-00007-of-00008.bin", "transformer.h.11.attn.c_attn.weight": "pytorch_model-00007-of-00008.bin", "transformer.h.11.attn.c_proj.bias": "pytorch_model-00007-of-00008.bin", "transformer.h.11.attn.c_proj.weight": "pytorch_model-00007-of-00008.bin", "transformer.h.11.attn.masked_bias": "pytorch_model-00007-of-00008.bin", "transformer.h.11.ln_1.bias": "pytorch_model-00007-of-00008.bin", "transformer.h.11.ln_1.weight": "pytorch_model-00007-of-00008.bin", "transformer.h.11.ln_2.bias": "pytorch_model-00007-of-00008.bin", "transformer.h.11.ln_2.weight": "pytorch_model-00007-of-00008.bin", "transformer.h.11.mlp.c_fc.bias": "pytorch_model-00007-of-00008.bin", "transformer.h.11.mlp.c_fc.weight": "pytorch_model-00007-of-00008.bin", "transformer.h.11.mlp.c_proj.bias": "pytorch_model-00007-of-00008.bin", "transformer.h.11.mlp.c_proj.weight": "pytorch_model-00007-of-00008.bin", "transformer.h.2.attn.bias": "pytorch_model-00004-of-00008.bin", "transformer.h.2.attn.c_attn.bias": "pytorch_model-00004-of-00008.bin", "transformer.h.2.attn.c_attn.weight": "pytorch_model-00004-of-00008.bin", "transformer.h.2.attn.c_proj.bias": "pytorch_model-00004-of-00008.bin", "transformer.h.2.attn.c_proj.weight": "pytorch_model-00004-of-00008.bin", "transformer.h.2.attn.masked_bias": "pytorch_model-00004-of-00008.bin", "transformer.h.2.ln_1.bias": "pytorch_model-00003-of-00008.bin", "transformer.h.2.ln_1.weight": "pytorch_model-00003-of-00008.bin", "transformer.h.2.ln_2.bias": "pytorch_model-00004-of-00008.bin", "transformer.h.2.ln_2.weight": "pytorch_model-00004-of-00008.bin", "transformer.h.2.mlp.c_fc.bias": "pytorch_model-00004-of-00008.bin", "transformer.h.2.mlp.c_fc.weight": "pytorch_model-00004-of-00008.bin", "transformer.h.2.mlp.c_proj.bias": "pytorch_model-00004-of-00008.bin", "transformer.h.2.mlp.c_proj.weight": "pytorch_model-00004-of-00008.bin", "transformer.h.3.attn.bias": "pytorch_model-00004-of-00008.bin", "transformer.h.3.attn.c_attn.bias": "pytorch_model-00004-of-00008.bin", "transformer.h.3.attn.c_attn.weight": "pytorch_model-00004-of-00008.bin", "transformer.h.3.attn.c_proj.bias": "pytorch_model-00004-of-00008.bin", "transformer.h.3.attn.c_proj.weight": "pytorch_model-00004-of-00008.bin", "transformer.h.3.attn.masked_bias": "pytorch_model-00004-of-00008.bin", "transformer.h.3.ln_1.bias": "pytorch_model-00004-of-00008.bin", "transformer.h.3.ln_1.weight": "pytorch_model-00004-of-00008.bin", "transformer.h.3.ln_2.bias": "pytorch_model-00004-of-00008.bin", "transformer.h.3.ln_2.weight": "pytorch_model-00004-of-00008.bin", "transformer.h.3.mlp.c_fc.bias": "pytorch_model-00004-of-00008.bin", "transformer.h.3.mlp.c_fc.weight": "pytorch_model-00004-of-00008.bin", "transformer.h.3.mlp.c_proj.bias": "pytorch_model-00004-of-00008.bin", "transformer.h.3.mlp.c_proj.weight": "pytorch_model-00004-of-00008.bin", "transformer.h.4.attn.bias": "pytorch_model-00004-of-00008.bin", "transformer.h.4.attn.c_attn.bias": "pytorch_model-00004-of-00008.bin", "transformer.h.4.attn.c_attn.weight": "pytorch_model-00004-of-00008.bin", "transformer.h.4.attn.c_proj.bias": "pytorch_model-00004-of-00008.bin", "transformer.h.4.attn.c_proj.weight": "pytorch_model-00004-of-00008.bin", "transformer.h.4.attn.masked_bias": "pytorch_model-00004-of-00008.bin", "transformer.h.4.ln_1.bias": "pytorch_model-00004-of-00008.bin", "transformer.h.4.ln_1.weight": "pytorch_model-00004-of-00008.bin", "transformer.h.4.ln_2.bias": "pytorch_model-00004-of-00008.bin", "transformer.h.4.ln_2.weight": "pytorch_model-00004-of-00008.bin", "transformer.h.4.mlp.c_fc.bias": "pytorch_model-00005-of-00008.bin", "transformer.h.4.mlp.c_fc.weight": "pytorch_model-00005-of-00008.bin", "transformer.h.4.mlp.c_proj.bias": "pytorch_model-00005-of-00008.bin", "transformer.h.4.mlp.c_proj.weight": "pytorch_model-00005-of-00008.bin", "transformer.h.5.attn.bias": "pytorch_model-00005-of-00008.bin", "transformer.h.5.attn.c_attn.bias": "pytorch_model-00005-of-00008.bin", "transformer.h.5.attn.c_attn.weight": "pytorch_model-00005-of-00008.bin", "transformer.h.5.attn.c_proj.bias": "pytorch_model-00005-of-00008.bin", "transformer.h.5.attn.c_proj.weight": "pytorch_model-00005-of-00008.bin", "transformer.h.5.attn.masked_bias": "pytorch_model-00005-of-00008.bin", "transformer.h.5.ln_1.bias": "pytorch_model-00005-of-00008.bin", "transformer.h.5.ln_1.weight": "pytorch_model-00005-of-00008.bin", "transformer.h.5.ln_2.bias": "pytorch_model-00005-of-00008.bin", "transformer.h.5.ln_2.weight": "pytorch_model-00005-of-00008.bin", "transformer.h.5.mlp.c_fc.bias": "pytorch_model-00005-of-00008.bin", "transformer.h.5.mlp.c_fc.weight": "pytorch_model-00005-of-00008.bin", "transformer.h.5.mlp.c_proj.bias": "pytorch_model-00005-of-00008.bin", "transformer.h.5.mlp.c_proj.weight": "pytorch_model-00005-of-00008.bin", "transformer.h.6.attn.bias": "pytorch_model-00005-of-00008.bin", "transformer.h.6.attn.c_attn.bias": "pytorch_model-00005-of-00008.bin", "transformer.h.6.attn.c_attn.weight": "pytorch_model-00005-of-00008.bin", "transformer.h.6.attn.c_proj.bias": "pytorch_model-00005-of-00008.bin", "transformer.h.6.attn.c_proj.weight": "pytorch_model-00005-of-00008.bin", "transformer.h.6.attn.masked_bias": "pytorch_model-00005-of-00008.bin", "transformer.h.6.ln_1.bias": "pytorch_model-00005-of-00008.bin", "transformer.h.6.ln_1.weight": "pytorch_model-00005-of-00008.bin", "transformer.h.6.ln_2.bias": "pytorch_model-00005-of-00008.bin", "transformer.h.6.ln_2.weight": "pytorch_model-00005-of-00008.bin", "transformer.h.6.mlp.c_fc.bias": "pytorch_model-00005-of-00008.bin", "transformer.h.6.mlp.c_fc.weight": "pytorch_model-00005-of-00008.bin", "transformer.h.6.mlp.c_proj.bias": "pytorch_model-00005-of-00008.bin", "transformer.h.6.mlp.c_proj.weight": "pytorch_model-00005-of-00008.bin", "transformer.h.7.attn.bias": "pytorch_model-00006-of-00008.bin", "transformer.h.7.attn.c_attn.bias": "pytorch_model-00006-of-00008.bin", "transformer.h.7.attn.c_attn.weight": "pytorch_model-00006-of-00008.bin", "transformer.h.7.attn.c_proj.bias": "pytorch_model-00006-of-00008.bin", "transformer.h.7.attn.c_proj.weight": "pytorch_model-00006-of-00008.bin", "transformer.h.7.attn.masked_bias": "pytorch_model-00006-of-00008.bin", "transformer.h.7.ln_1.bias": "pytorch_model-00005-of-00008.bin", "transformer.h.7.ln_1.weight": "pytorch_model-00005-of-00008.bin", "transformer.h.7.ln_2.bias": "pytorch_model-00006-of-00008.bin", "transformer.h.7.ln_2.weight": "pytorch_model-00006-of-00008.bin", "transformer.h.7.mlp.c_fc.bias": "pytorch_model-00006-of-00008.bin", "transformer.h.7.mlp.c_fc.weight": "pytorch_model-00006-of-00008.bin", "transformer.h.7.mlp.c_proj.bias": "pytorch_model-00006-of-00008.bin", "transformer.h.7.mlp.c_proj.weight": "pytorch_model-00006-of-00008.bin", "transformer.h.8.attn.bias": "pytorch_model-00006-of-00008.bin", "transformer.h.8.attn.c_attn.bias": "pytorch_model-00006-of-00008.bin", "transformer.h.8.attn.c_attn.weight": "pytorch_model-00006-of-00008.bin", "transformer.h.8.attn.c_proj.bias": "pytorch_model-00006-of-00008.bin", "transformer.h.8.attn.c_proj.weight": "pytorch_model-00006-of-00008.bin", "transformer.h.8.attn.masked_bias": "pytorch_model-00006-of-00008.bin", "transformer.h.8.ln_1.bias": "pytorch_model-00006-of-00008.bin", "transformer.h.8.ln_1.weight": "pytorch_model-00006-of-00008.bin", "transformer.h.8.ln_2.bias": "pytorch_model-00006-of-00008.bin", "transformer.h.8.ln_2.weight": "pytorch_model-00006-of-00008.bin", "transformer.h.8.mlp.c_fc.bias": "pytorch_model-00006-of-00008.bin", "transformer.h.8.mlp.c_fc.weight": "pytorch_model-00006-of-00008.bin", "transformer.h.8.mlp.c_proj.bias": "pytorch_model-00006-of-00008.bin", "transformer.h.8.mlp.c_proj.weight": "pytorch_model-00006-of-00008.bin", "transformer.h.9.attn.bias": "pytorch_model-00006-of-00008.bin", "transformer.h.9.attn.c_attn.bias": "pytorch_model-00006-of-00008.bin", "transformer.h.9.attn.c_attn.weight": "pytorch_model-00006-of-00008.bin", "transformer.h.9.attn.c_proj.bias": "pytorch_model-00006-of-00008.bin", "transformer.h.9.attn.c_proj.weight": "pytorch_model-00006-of-00008.bin", "transformer.h.9.attn.masked_bias": "pytorch_model-00006-of-00008.bin", "transformer.h.9.ln_1.bias": "pytorch_model-00006-of-00008.bin", "transformer.h.9.ln_1.weight": "pytorch_model-00006-of-00008.bin", "transformer.h.9.ln_2.bias": "pytorch_model-00006-of-00008.bin", "transformer.h.9.ln_2.weight": "pytorch_model-00006-of-00008.bin", "transformer.h.9.mlp.c_fc.bias": "pytorch_model-00007-of-00008.bin", "transformer.h.9.mlp.c_fc.weight": "pytorch_model-00007-of-00008.bin", "transformer.h.9.mlp.c_proj.bias": "pytorch_model-00007-of-00008.bin", "transformer.h.9.mlp.c_proj.weight": "pytorch_model-00007-of-00008.bin", "transformer.ln_f.bias": "pytorch_model-00007-of-00008.bin", "transformer.ln_f.weight": "pytorch_model-00007-of-00008.bin", "transformer.wpe.weight": "pytorch_model-00003-of-00008.bin", "transformer.wte.weight": "pytorch_model-00002-of-00008.bin" } }