{ "bos_token_id": 1, "eos_token_id": 2, "hidden_act": "silu", "hidden_size": 4096, "initializer_range": 0.02, "intermediate_size": 11008, "max_position_embeddings": 4096, "model": "llama", "num_attention_heads": 32, "num_hidden_layers": 32, "multi_query_group_num": 4, "pad_token_id": 0, "layer_norm_eps": 1e-05, "rope_theta": 5000000.0, "tie_word_embeddings": false, "torch_dtype": "bfloat16", "vocab_size": 64000, "skip_init": true, "rope_rank": "updown", "segment_vocab_size": 0, "generation_config": {"tokenizer_decode_config": {"skip_special_tokens": true}, "max_length": 4096, "end_id": 2}, "mapping": { "embeddings.word_embeddings.weight": "model.embed_tokens.weight", "LayerNormFinal.weight": "model.norm.weight", "lm_head.weight": "lm_head.weight", "decoderLayer.0.multiHeadAttention.q.weight": "model.layers.0.self_attn.q_proj.weight", "decoderLayer.0.multiHeadAttention.k.weight": "model.layers.0.self_attn.k_proj.weight", "decoderLayer.0.multiHeadAttention.v.weight": "model.layers.0.self_attn.v_proj.weight", "decoderLayer.0.multiHeadAttention.o.weight": "model.layers.0.self_attn.o_proj.weight", "decoderLayer.0.attnLayerNorm.weight": "model.layers.0.ln1.weight", "decoderLayer.0.feedForward.intermediateDense.weight": "model.layers.0.mlp.gate_proj.weight", "decoderLayer.0.feedForward.outputDense.weight": "model.layers.0.mlp.down_proj.weight", "decoderLayer.0.ffnLayerNorm.weight": "model.layers.0.ln2.weight", "decoderLayer.0.feedForward.intermediateDense2.weight": "model.layers.0.mlp.up_proj.weight", "decoderLayer.1.multiHeadAttention.q.weight": "model.layers.1.self_attn.q_proj.weight", "decoderLayer.1.multiHeadAttention.k.weight": "model.layers.1.self_attn.k_proj.weight", "decoderLayer.1.multiHeadAttention.v.weight": "model.layers.1.self_attn.v_proj.weight", "decoderLayer.1.multiHeadAttention.o.weight": "model.layers.1.self_attn.o_proj.weight", "decoderLayer.1.attnLayerNorm.weight": "model.layers.1.ln1.weight", "decoderLayer.1.feedForward.intermediateDense.weight": "model.layers.1.mlp.gate_proj.weight", "decoderLayer.1.feedForward.outputDense.weight": "model.layers.1.mlp.down_proj.weight", "decoderLayer.1.ffnLayerNorm.weight": "model.layers.1.ln2.weight", "decoderLayer.1.feedForward.intermediateDense2.weight": "model.layers.1.mlp.up_proj.weight", "decoderLayer.2.multiHeadAttention.q.weight": "model.layers.2.self_attn.q_proj.weight", "decoderLayer.2.multiHeadAttention.k.weight": "model.layers.2.self_attn.k_proj.weight", "decoderLayer.2.multiHeadAttention.v.weight": "model.layers.2.self_attn.v_proj.weight", "decoderLayer.2.multiHeadAttention.o.weight": "model.layers.2.self_attn.o_proj.weight", "decoderLayer.2.attnLayerNorm.weight": "model.layers.2.ln1.weight", "decoderLayer.2.feedForward.intermediateDense.weight": "model.layers.2.mlp.gate_proj.weight", "decoderLayer.2.feedForward.outputDense.weight": "model.layers.2.mlp.down_proj.weight", "decoderLayer.2.ffnLayerNorm.weight": "model.layers.2.ln2.weight", "decoderLayer.2.feedForward.intermediateDense2.weight": "model.layers.2.mlp.up_proj.weight", "decoderLayer.3.multiHeadAttention.q.weight": "model.layers.3.self_attn.q_proj.weight", "decoderLayer.3.multiHeadAttention.k.weight": "model.layers.3.self_attn.k_proj.weight", "decoderLayer.3.multiHeadAttention.v.weight": "model.layers.3.self_attn.v_proj.weight", "decoderLayer.3.multiHeadAttention.o.weight": "model.layers.3.self_attn.o_proj.weight", "decoderLayer.3.attnLayerNorm.weight": "model.layers.3.ln1.weight", "decoderLayer.3.feedForward.intermediateDense.weight": "model.layers.3.mlp.gate_proj.weight", "decoderLayer.3.feedForward.outputDense.weight": "model.layers.3.mlp.down_proj.weight", "decoderLayer.3.ffnLayerNorm.weight": "model.layers.3.ln2.weight", "decoderLayer.3.feedForward.intermediateDense2.weight": "model.layers.3.mlp.up_proj.weight", "decoderLayer.4.multiHeadAttention.q.weight": "model.layers.4.self_attn.q_proj.weight", "decoderLayer.4.multiHeadAttention.k.weight": "model.layers.4.self_attn.k_proj.weight", "decoderLayer.4.multiHeadAttention.v.weight": "model.layers.4.self_attn.v_proj.weight", "decoderLayer.4.multiHeadAttention.o.weight": "model.layers.4.self_attn.o_proj.weight", "decoderLayer.4.attnLayerNorm.weight": "model.layers.4.ln1.weight", "decoderLayer.4.feedForward.intermediateDense.weight": "model.layers.4.mlp.gate_proj.weight", "decoderLayer.4.feedForward.outputDense.weight": "model.layers.4.mlp.down_proj.weight", "decoderLayer.4.ffnLayerNorm.weight": "model.layers.4.ln2.weight", "decoderLayer.4.feedForward.intermediateDense2.weight": "model.layers.4.mlp.up_proj.weight", "decoderLayer.5.multiHeadAttention.q.weight": "model.layers.5.self_attn.q_proj.weight", "decoderLayer.5.multiHeadAttention.k.weight": "model.layers.5.self_attn.k_proj.weight", "decoderLayer.5.multiHeadAttention.v.weight": "model.layers.5.self_attn.v_proj.weight", "decoderLayer.5.multiHeadAttention.o.weight": "model.layers.5.self_attn.o_proj.weight", "decoderLayer.5.attnLayerNorm.weight": "model.layers.5.ln1.weight", "decoderLayer.5.feedForward.intermediateDense.weight": "model.layers.5.mlp.gate_proj.weight", "decoderLayer.5.feedForward.outputDense.weight": "model.layers.5.mlp.down_proj.weight", "decoderLayer.5.ffnLayerNorm.weight": "model.layers.5.ln2.weight", "decoderLayer.5.feedForward.intermediateDense2.weight": "model.layers.5.mlp.up_proj.weight", "decoderLayer.6.multiHeadAttention.q.weight": "model.layers.6.self_attn.q_proj.weight", "decoderLayer.6.multiHeadAttention.k.weight": "model.layers.6.self_attn.k_proj.weight", "decoderLayer.6.multiHeadAttention.v.weight": "model.layers.6.self_attn.v_proj.weight", "decoderLayer.6.multiHeadAttention.o.weight": "model.layers.6.self_attn.o_proj.weight", "decoderLayer.6.attnLayerNorm.weight": "model.layers.6.ln1.weight", "decoderLayer.6.feedForward.intermediateDense.weight": "model.layers.6.mlp.gate_proj.weight", "decoderLayer.6.feedForward.outputDense.weight": "model.layers.6.mlp.down_proj.weight", "decoderLayer.6.ffnLayerNorm.weight": "model.layers.6.ln2.weight", "decoderLayer.6.feedForward.intermediateDense2.weight": "model.layers.6.mlp.up_proj.weight", "decoderLayer.7.multiHeadAttention.q.weight": "model.layers.7.self_attn.q_proj.weight", "decoderLayer.7.multiHeadAttention.k.weight": "model.layers.7.self_attn.k_proj.weight", "decoderLayer.7.multiHeadAttention.v.weight": "model.layers.7.self_attn.v_proj.weight", "decoderLayer.7.multiHeadAttention.o.weight": "model.layers.7.self_attn.o_proj.weight", "decoderLayer.7.attnLayerNorm.weight": "model.layers.7.ln1.weight", "decoderLayer.7.feedForward.intermediateDense.weight": "model.layers.7.mlp.gate_proj.weight", "decoderLayer.7.feedForward.outputDense.weight": "model.layers.7.mlp.down_proj.weight", "decoderLayer.7.ffnLayerNorm.weight": "model.layers.7.ln2.weight", "decoderLayer.7.feedForward.intermediateDense2.weight": "model.layers.7.mlp.up_proj.weight", "decoderLayer.8.multiHeadAttention.q.weight": "model.layers.8.self_attn.q_proj.weight", "decoderLayer.8.multiHeadAttention.k.weight": "model.layers.8.self_attn.k_proj.weight", "decoderLayer.8.multiHeadAttention.v.weight": "model.layers.8.self_attn.v_proj.weight", "decoderLayer.8.multiHeadAttention.o.weight": "model.layers.8.self_attn.o_proj.weight", "decoderLayer.8.attnLayerNorm.weight": "model.layers.8.ln1.weight", "decoderLayer.8.feedForward.intermediateDense.weight": "model.layers.8.mlp.gate_proj.weight", "decoderLayer.8.feedForward.outputDense.weight": "model.layers.8.mlp.down_proj.weight", "decoderLayer.8.ffnLayerNorm.weight": "model.layers.8.ln2.weight", "decoderLayer.8.feedForward.intermediateDense2.weight": "model.layers.8.mlp.up_proj.weight", "decoderLayer.9.multiHeadAttention.q.weight": "model.layers.9.self_attn.q_proj.weight", "decoderLayer.9.multiHeadAttention.k.weight": "model.layers.9.self_attn.k_proj.weight", "decoderLayer.9.multiHeadAttention.v.weight": "model.layers.9.self_attn.v_proj.weight", "decoderLayer.9.multiHeadAttention.o.weight": "model.layers.9.self_attn.o_proj.weight", "decoderLayer.9.attnLayerNorm.weight": "model.layers.9.ln1.weight", "decoderLayer.9.feedForward.intermediateDense.weight": "model.layers.9.mlp.gate_proj.weight", "decoderLayer.9.feedForward.outputDense.weight": "model.layers.9.mlp.down_proj.weight", "decoderLayer.9.ffnLayerNorm.weight": "model.layers.9.ln2.weight", "decoderLayer.9.feedForward.intermediateDense2.weight": "model.layers.9.mlp.up_proj.weight", "decoderLayer.10.multiHeadAttention.q.weight": "model.layers.10.self_attn.q_proj.weight", "decoderLayer.10.multiHeadAttention.k.weight": "model.layers.10.self_attn.k_proj.weight", "decoderLayer.10.multiHeadAttention.v.weight": "model.layers.10.self_attn.v_proj.weight", "decoderLayer.10.multiHeadAttention.o.weight": "model.layers.10.self_attn.o_proj.weight", "decoderLayer.10.attnLayerNorm.weight": "model.layers.10.ln1.weight", "decoderLayer.10.feedForward.intermediateDense.weight": "model.layers.10.mlp.gate_proj.weight", "decoderLayer.10.feedForward.outputDense.weight": "model.layers.10.mlp.down_proj.weight", "decoderLayer.10.ffnLayerNorm.weight": "model.layers.10.ln2.weight", "decoderLayer.10.feedForward.intermediateDense2.weight": "model.layers.10.mlp.up_proj.weight", "decoderLayer.11.multiHeadAttention.q.weight": "model.layers.11.self_attn.q_proj.weight", "decoderLayer.11.multiHeadAttention.k.weight": "model.layers.11.self_attn.k_proj.weight", "decoderLayer.11.multiHeadAttention.v.weight": "model.layers.11.self_attn.v_proj.weight", "decoderLayer.11.multiHeadAttention.o.weight": "model.layers.11.self_attn.o_proj.weight", "decoderLayer.11.attnLayerNorm.weight": "model.layers.11.ln1.weight", "decoderLayer.11.feedForward.intermediateDense.weight": "model.layers.11.mlp.gate_proj.weight", "decoderLayer.11.feedForward.outputDense.weight": "model.layers.11.mlp.down_proj.weight", "decoderLayer.11.ffnLayerNorm.weight": "model.layers.11.ln2.weight", "decoderLayer.11.feedForward.intermediateDense2.weight": "model.layers.11.mlp.up_proj.weight", "decoderLayer.12.multiHeadAttention.q.weight": "model.layers.12.self_attn.q_proj.weight", "decoderLayer.12.multiHeadAttention.k.weight": "model.layers.12.self_attn.k_proj.weight", "decoderLayer.12.multiHeadAttention.v.weight": "model.layers.12.self_attn.v_proj.weight", "decoderLayer.12.multiHeadAttention.o.weight": "model.layers.12.self_attn.o_proj.weight", "decoderLayer.12.attnLayerNorm.weight": "model.layers.12.ln1.weight", "decoderLayer.12.feedForward.intermediateDense.weight": "model.layers.12.mlp.gate_proj.weight", "decoderLayer.12.feedForward.outputDense.weight": "model.layers.12.mlp.down_proj.weight", "decoderLayer.12.ffnLayerNorm.weight": "model.layers.12.ln2.weight", "decoderLayer.12.feedForward.intermediateDense2.weight": "model.layers.12.mlp.up_proj.weight", "decoderLayer.13.multiHeadAttention.q.weight": "model.layers.13.self_attn.q_proj.weight", "decoderLayer.13.multiHeadAttention.k.weight": "model.layers.13.self_attn.k_proj.weight", "decoderLayer.13.multiHeadAttention.v.weight": "model.layers.13.self_attn.v_proj.weight", "decoderLayer.13.multiHeadAttention.o.weight": "model.layers.13.self_attn.o_proj.weight", "decoderLayer.13.attnLayerNorm.weight": "model.layers.13.ln1.weight", "decoderLayer.13.feedForward.intermediateDense.weight": "model.layers.13.mlp.gate_proj.weight", "decoderLayer.13.feedForward.outputDense.weight": "model.layers.13.mlp.down_proj.weight", "decoderLayer.13.ffnLayerNorm.weight": "model.layers.13.ln2.weight", "decoderLayer.13.feedForward.intermediateDense2.weight": "model.layers.13.mlp.up_proj.weight", "decoderLayer.14.multiHeadAttention.q.weight": "model.layers.14.self_attn.q_proj.weight", "decoderLayer.14.multiHeadAttention.k.weight": "model.layers.14.self_attn.k_proj.weight", "decoderLayer.14.multiHeadAttention.v.weight": "model.layers.14.self_attn.v_proj.weight", "decoderLayer.14.multiHeadAttention.o.weight": "model.layers.14.self_attn.o_proj.weight", "decoderLayer.14.attnLayerNorm.weight": "model.layers.14.ln1.weight", "decoderLayer.14.feedForward.intermediateDense.weight": "model.layers.14.mlp.gate_proj.weight", "decoderLayer.14.feedForward.outputDense.weight": "model.layers.14.mlp.down_proj.weight", "decoderLayer.14.ffnLayerNorm.weight": "model.layers.14.ln2.weight", "decoderLayer.14.feedForward.intermediateDense2.weight": "model.layers.14.mlp.up_proj.weight", "decoderLayer.15.multiHeadAttention.q.weight": "model.layers.15.self_attn.q_proj.weight", "decoderLayer.15.multiHeadAttention.k.weight": "model.layers.15.self_attn.k_proj.weight", "decoderLayer.15.multiHeadAttention.v.weight": "model.layers.15.self_attn.v_proj.weight", "decoderLayer.15.multiHeadAttention.o.weight": "model.layers.15.self_attn.o_proj.weight", "decoderLayer.15.attnLayerNorm.weight": "model.layers.15.ln1.weight", "decoderLayer.15.feedForward.intermediateDense.weight": "model.layers.15.mlp.gate_proj.weight", "decoderLayer.15.feedForward.outputDense.weight": "model.layers.15.mlp.down_proj.weight", "decoderLayer.15.ffnLayerNorm.weight": "model.layers.15.ln2.weight", "decoderLayer.15.feedForward.intermediateDense2.weight": "model.layers.15.mlp.up_proj.weight", "decoderLayer.16.multiHeadAttention.q.weight": "model.layers.16.self_attn.q_proj.weight", "decoderLayer.16.multiHeadAttention.k.weight": "model.layers.16.self_attn.k_proj.weight", "decoderLayer.16.multiHeadAttention.v.weight": "model.layers.16.self_attn.v_proj.weight", "decoderLayer.16.multiHeadAttention.o.weight": "model.layers.16.self_attn.o_proj.weight", "decoderLayer.16.attnLayerNorm.weight": "model.layers.16.ln1.weight", "decoderLayer.16.feedForward.intermediateDense.weight": "model.layers.16.mlp.gate_proj.weight", "decoderLayer.16.feedForward.outputDense.weight": "model.layers.16.mlp.down_proj.weight", "decoderLayer.16.ffnLayerNorm.weight": "model.layers.16.ln2.weight", "decoderLayer.16.feedForward.intermediateDense2.weight": "model.layers.16.mlp.up_proj.weight", "decoderLayer.17.multiHeadAttention.q.weight": "model.layers.17.self_attn.q_proj.weight", "decoderLayer.17.multiHeadAttention.k.weight": "model.layers.17.self_attn.k_proj.weight", "decoderLayer.17.multiHeadAttention.v.weight": "model.layers.17.self_attn.v_proj.weight", "decoderLayer.17.multiHeadAttention.o.weight": "model.layers.17.self_attn.o_proj.weight", "decoderLayer.17.attnLayerNorm.weight": "model.layers.17.ln1.weight", "decoderLayer.17.feedForward.intermediateDense.weight": "model.layers.17.mlp.gate_proj.weight", "decoderLayer.17.feedForward.outputDense.weight": "model.layers.17.mlp.down_proj.weight", "decoderLayer.17.ffnLayerNorm.weight": "model.layers.17.ln2.weight", "decoderLayer.17.feedForward.intermediateDense2.weight": "model.layers.17.mlp.up_proj.weight", "decoderLayer.18.multiHeadAttention.q.weight": "model.layers.18.self_attn.q_proj.weight", "decoderLayer.18.multiHeadAttention.k.weight": "model.layers.18.self_attn.k_proj.weight", "decoderLayer.18.multiHeadAttention.v.weight": "model.layers.18.self_attn.v_proj.weight", "decoderLayer.18.multiHeadAttention.o.weight": "model.layers.18.self_attn.o_proj.weight", "decoderLayer.18.attnLayerNorm.weight": "model.layers.18.ln1.weight", "decoderLayer.18.feedForward.intermediateDense.weight": "model.layers.18.mlp.gate_proj.weight", "decoderLayer.18.feedForward.outputDense.weight": "model.layers.18.mlp.down_proj.weight", "decoderLayer.18.ffnLayerNorm.weight": "model.layers.18.ln2.weight", "decoderLayer.18.feedForward.intermediateDense2.weight": "model.layers.18.mlp.up_proj.weight", "decoderLayer.19.multiHeadAttention.q.weight": "model.layers.19.self_attn.q_proj.weight", "decoderLayer.19.multiHeadAttention.k.weight": "model.layers.19.self_attn.k_proj.weight", "decoderLayer.19.multiHeadAttention.v.weight": "model.layers.19.self_attn.v_proj.weight", "decoderLayer.19.multiHeadAttention.o.weight": "model.layers.19.self_attn.o_proj.weight", "decoderLayer.19.attnLayerNorm.weight": "model.layers.19.ln1.weight", "decoderLayer.19.feedForward.intermediateDense.weight": "model.layers.19.mlp.gate_proj.weight", "decoderLayer.19.feedForward.outputDense.weight": "model.layers.19.mlp.down_proj.weight", "decoderLayer.19.ffnLayerNorm.weight": "model.layers.19.ln2.weight", "decoderLayer.19.feedForward.intermediateDense2.weight": "model.layers.19.mlp.up_proj.weight", "decoderLayer.20.multiHeadAttention.q.weight": "model.layers.20.self_attn.q_proj.weight", "decoderLayer.20.multiHeadAttention.k.weight": "model.layers.20.self_attn.k_proj.weight", "decoderLayer.20.multiHeadAttention.v.weight": "model.layers.20.self_attn.v_proj.weight", "decoderLayer.20.multiHeadAttention.o.weight": "model.layers.20.self_attn.o_proj.weight", "decoderLayer.20.attnLayerNorm.weight": "model.layers.20.ln1.weight", "decoderLayer.20.feedForward.intermediateDense.weight": "model.layers.20.mlp.gate_proj.weight", "decoderLayer.20.feedForward.outputDense.weight": "model.layers.20.mlp.down_proj.weight", "decoderLayer.20.ffnLayerNorm.weight": "model.layers.20.ln2.weight", "decoderLayer.20.feedForward.intermediateDense2.weight": "model.layers.20.mlp.up_proj.weight", "decoderLayer.21.multiHeadAttention.q.weight": "model.layers.21.self_attn.q_proj.weight", "decoderLayer.21.multiHeadAttention.k.weight": "model.layers.21.self_attn.k_proj.weight", "decoderLayer.21.multiHeadAttention.v.weight": "model.layers.21.self_attn.v_proj.weight", "decoderLayer.21.multiHeadAttention.o.weight": "model.layers.21.self_attn.o_proj.weight", "decoderLayer.21.attnLayerNorm.weight": "model.layers.21.ln1.weight", "decoderLayer.21.feedForward.intermediateDense.weight": "model.layers.21.mlp.gate_proj.weight", "decoderLayer.21.feedForward.outputDense.weight": "model.layers.21.mlp.down_proj.weight", "decoderLayer.21.ffnLayerNorm.weight": "model.layers.21.ln2.weight", "decoderLayer.21.feedForward.intermediateDense2.weight": "model.layers.21.mlp.up_proj.weight", "decoderLayer.22.multiHeadAttention.q.weight": "model.layers.22.self_attn.q_proj.weight", "decoderLayer.22.multiHeadAttention.k.weight": "model.layers.22.self_attn.k_proj.weight", "decoderLayer.22.multiHeadAttention.v.weight": "model.layers.22.self_attn.v_proj.weight", "decoderLayer.22.multiHeadAttention.o.weight": "model.layers.22.self_attn.o_proj.weight", "decoderLayer.22.attnLayerNorm.weight": "model.layers.22.ln1.weight", "decoderLayer.22.feedForward.intermediateDense.weight": "model.layers.22.mlp.gate_proj.weight", "decoderLayer.22.feedForward.outputDense.weight": "model.layers.22.mlp.down_proj.weight", "decoderLayer.22.ffnLayerNorm.weight": "model.layers.22.ln2.weight", "decoderLayer.22.feedForward.intermediateDense2.weight": "model.layers.22.mlp.up_proj.weight", "decoderLayer.23.multiHeadAttention.q.weight": "model.layers.23.self_attn.q_proj.weight", "decoderLayer.23.multiHeadAttention.k.weight": "model.layers.23.self_attn.k_proj.weight", "decoderLayer.23.multiHeadAttention.v.weight": "model.layers.23.self_attn.v_proj.weight", "decoderLayer.23.multiHeadAttention.o.weight": "model.layers.23.self_attn.o_proj.weight", "decoderLayer.23.attnLayerNorm.weight": "model.layers.23.ln1.weight", "decoderLayer.23.feedForward.intermediateDense.weight": "model.layers.23.mlp.gate_proj.weight", "decoderLayer.23.feedForward.outputDense.weight": "model.layers.23.mlp.down_proj.weight", "decoderLayer.23.ffnLayerNorm.weight": "model.layers.23.ln2.weight", "decoderLayer.23.feedForward.intermediateDense2.weight": "model.layers.23.mlp.up_proj.weight", "decoderLayer.24.multiHeadAttention.q.weight": "model.layers.24.self_attn.q_proj.weight", "decoderLayer.24.multiHeadAttention.k.weight": "model.layers.24.self_attn.k_proj.weight", "decoderLayer.24.multiHeadAttention.v.weight": "model.layers.24.self_attn.v_proj.weight", "decoderLayer.24.multiHeadAttention.o.weight": "model.layers.24.self_attn.o_proj.weight", "decoderLayer.24.attnLayerNorm.weight": "model.layers.24.ln1.weight", "decoderLayer.24.feedForward.intermediateDense.weight": "model.layers.24.mlp.gate_proj.weight", "decoderLayer.24.feedForward.outputDense.weight": "model.layers.24.mlp.down_proj.weight", "decoderLayer.24.ffnLayerNorm.weight": "model.layers.24.ln2.weight", "decoderLayer.24.feedForward.intermediateDense2.weight": "model.layers.24.mlp.up_proj.weight", "decoderLayer.25.multiHeadAttention.q.weight": "model.layers.25.self_attn.q_proj.weight", "decoderLayer.25.multiHeadAttention.k.weight": "model.layers.25.self_attn.k_proj.weight", "decoderLayer.25.multiHeadAttention.v.weight": "model.layers.25.self_attn.v_proj.weight", "decoderLayer.25.multiHeadAttention.o.weight": "model.layers.25.self_attn.o_proj.weight", "decoderLayer.25.attnLayerNorm.weight": "model.layers.25.ln1.weight", "decoderLayer.25.feedForward.intermediateDense.weight": "model.layers.25.mlp.gate_proj.weight", "decoderLayer.25.feedForward.outputDense.weight": "model.layers.25.mlp.down_proj.weight", "decoderLayer.25.ffnLayerNorm.weight": "model.layers.25.ln2.weight", "decoderLayer.25.feedForward.intermediateDense2.weight": "model.layers.25.mlp.up_proj.weight", "decoderLayer.26.multiHeadAttention.q.weight": "model.layers.26.self_attn.q_proj.weight", "decoderLayer.26.multiHeadAttention.k.weight": "model.layers.26.self_attn.k_proj.weight", "decoderLayer.26.multiHeadAttention.v.weight": "model.layers.26.self_attn.v_proj.weight", "decoderLayer.26.multiHeadAttention.o.weight": "model.layers.26.self_attn.o_proj.weight", "decoderLayer.26.attnLayerNorm.weight": "model.layers.26.ln1.weight", "decoderLayer.26.feedForward.intermediateDense.weight": "model.layers.26.mlp.gate_proj.weight", "decoderLayer.26.feedForward.outputDense.weight": "model.layers.26.mlp.down_proj.weight", "decoderLayer.26.ffnLayerNorm.weight": "model.layers.26.ln2.weight", "decoderLayer.26.feedForward.intermediateDense2.weight": "model.layers.26.mlp.up_proj.weight", "decoderLayer.27.multiHeadAttention.q.weight": "model.layers.27.self_attn.q_proj.weight", "decoderLayer.27.multiHeadAttention.k.weight": "model.layers.27.self_attn.k_proj.weight", "decoderLayer.27.multiHeadAttention.v.weight": "model.layers.27.self_attn.v_proj.weight", "decoderLayer.27.multiHeadAttention.o.weight": "model.layers.27.self_attn.o_proj.weight", "decoderLayer.27.attnLayerNorm.weight": "model.layers.27.ln1.weight", "decoderLayer.27.feedForward.intermediateDense.weight": "model.layers.27.mlp.gate_proj.weight", "decoderLayer.27.feedForward.outputDense.weight": "model.layers.27.mlp.down_proj.weight", "decoderLayer.27.ffnLayerNorm.weight": "model.layers.27.ln2.weight", "decoderLayer.27.feedForward.intermediateDense2.weight": "model.layers.27.mlp.up_proj.weight", "decoderLayer.28.multiHeadAttention.q.weight": "model.layers.28.self_attn.q_proj.weight", "decoderLayer.28.multiHeadAttention.k.weight": "model.layers.28.self_attn.k_proj.weight", "decoderLayer.28.multiHeadAttention.v.weight": "model.layers.28.self_attn.v_proj.weight", "decoderLayer.28.multiHeadAttention.o.weight": "model.layers.28.self_attn.o_proj.weight", "decoderLayer.28.attnLayerNorm.weight": "model.layers.28.ln1.weight", "decoderLayer.28.feedForward.intermediateDense.weight": "model.layers.28.mlp.gate_proj.weight", "decoderLayer.28.feedForward.outputDense.weight": "model.layers.28.mlp.down_proj.weight", "decoderLayer.28.ffnLayerNorm.weight": "model.layers.28.ln2.weight", "decoderLayer.28.feedForward.intermediateDense2.weight": "model.layers.28.mlp.up_proj.weight", "decoderLayer.29.multiHeadAttention.q.weight": "model.layers.29.self_attn.q_proj.weight", "decoderLayer.29.multiHeadAttention.k.weight": "model.layers.29.self_attn.k_proj.weight", "decoderLayer.29.multiHeadAttention.v.weight": "model.layers.29.self_attn.v_proj.weight", "decoderLayer.29.multiHeadAttention.o.weight": "model.layers.29.self_attn.o_proj.weight", "decoderLayer.29.attnLayerNorm.weight": "model.layers.29.ln1.weight", "decoderLayer.29.feedForward.intermediateDense.weight": "model.layers.29.mlp.gate_proj.weight", "decoderLayer.29.feedForward.outputDense.weight": "model.layers.29.mlp.down_proj.weight", "decoderLayer.29.ffnLayerNorm.weight": "model.layers.29.ln2.weight", "decoderLayer.29.feedForward.intermediateDense2.weight": "model.layers.29.mlp.up_proj.weight", "decoderLayer.30.multiHeadAttention.q.weight": "model.layers.30.self_attn.q_proj.weight", "decoderLayer.30.multiHeadAttention.k.weight": "model.layers.30.self_attn.k_proj.weight", "decoderLayer.30.multiHeadAttention.v.weight": "model.layers.30.self_attn.v_proj.weight", "decoderLayer.30.multiHeadAttention.o.weight": "model.layers.30.self_attn.o_proj.weight", "decoderLayer.30.attnLayerNorm.weight": "model.layers.30.ln1.weight", "decoderLayer.30.feedForward.intermediateDense.weight": "model.layers.30.mlp.gate_proj.weight", "decoderLayer.30.feedForward.outputDense.weight": "model.layers.30.mlp.down_proj.weight", "decoderLayer.30.ffnLayerNorm.weight": "model.layers.30.ln2.weight", "decoderLayer.30.feedForward.intermediateDense2.weight": "model.layers.30.mlp.up_proj.weight", "decoderLayer.31.multiHeadAttention.q.weight": "model.layers.31.self_attn.q_proj.weight", "decoderLayer.31.multiHeadAttention.k.weight": "model.layers.31.self_attn.k_proj.weight", "decoderLayer.31.multiHeadAttention.v.weight": "model.layers.31.self_attn.v_proj.weight", "decoderLayer.31.multiHeadAttention.o.weight": "model.layers.31.self_attn.o_proj.weight", "decoderLayer.31.attnLayerNorm.weight": "model.layers.31.ln1.weight", "decoderLayer.31.feedForward.intermediateDense.weight": "model.layers.31.mlp.gate_proj.weight", "decoderLayer.31.feedForward.outputDense.weight": "model.layers.31.mlp.down_proj.weight", "decoderLayer.31.ffnLayerNorm.weight": "model.layers.31.ln2.weight", "decoderLayer.31.feedForward.intermediateDense2.weight": "model.layers.31.mlp.up_proj.weight" } }