load_keys.json

{"all_checkpoint_keys": ["transformer.word_embeddings.weight", "transformer.word_embeddings_layernorm.weight", "transformer.word_embeddings_layernorm.bias", "transformer.h.0.input_layernorm.weight", "transformer.h.0.input_layernorm.bias", "transformer.h.0.self_attention.query_key_value.weight", "transformer.h.0.self_attention.query_key_value.bias", "transformer.h.0.self_attention.dense.weight", "transformer.h.0.self_attention.dense.bias", "transformer.h.0.post_attention_layernorm.weight", "transformer.h.0.post_attention_layernorm.bias", "transformer.h.0.mlp.dense_h_to_4h.weight", "transformer.h.0.mlp.dense_h_to_4h.bias", "transformer.h.0.mlp.dense_4h_to_h.weight", "transformer.h.0.mlp.dense_4h_to_h.bias", "transformer.h.1.input_layernorm.weight", "transformer.h.1.input_layernorm.bias", "transformer.h.1.self_attention.query_key_value.weight", "transformer.h.1.self_attention.query_key_value.bias", "transformer.h.1.self_attention.dense.weight", "transformer.h.1.self_attention.dense.bias", "transformer.h.1.post_attention_layernorm.weight", "transformer.h.1.post_attention_layernorm.bias", "transformer.h.1.mlp.dense_h_to_4h.weight", "transformer.h.1.mlp.dense_h_to_4h.bias", "transformer.h.1.mlp.dense_4h_to_h.weight", "transformer.h.1.mlp.dense_4h_to_h.bias", "transformer.h.2.input_layernorm.weight", "transformer.h.2.input_layernorm.bias", "transformer.h.2.self_attention.query_key_value.weight", "transformer.h.2.self_attention.query_key_value.bias", "transformer.h.2.self_attention.dense.weight", "transformer.h.2.self_attention.dense.bias", "transformer.h.2.post_attention_layernorm.weight", "transformer.h.2.post_attention_layernorm.bias", "transformer.h.2.mlp.dense_h_to_4h.weight", "transformer.h.2.mlp.dense_h_to_4h.bias", "transformer.h.2.mlp.dense_4h_to_h.weight", "transformer.h.2.mlp.dense_4h_to_h.bias", "transformer.h.3.input_layernorm.weight", "transformer.h.3.input_layernorm.bias", "transformer.h.3.self_attention.query_key_value.weight", "transformer.h.3.self_attention.query_key_value.bias", "transformer.h.3.self_attention.dense.weight", "transformer.h.3.self_attention.dense.bias", "transformer.h.3.post_attention_layernorm.weight", "transformer.h.3.post_attention_layernorm.bias", "transformer.h.3.mlp.dense_h_to_4h.weight", "transformer.h.3.mlp.dense_h_to_4h.bias", "transformer.h.3.mlp.dense_4h_to_h.weight", "transformer.h.3.mlp.dense_4h_to_h.bias", "transformer.h.4.input_layernorm.weight", "transformer.h.4.input_layernorm.bias", "transformer.h.4.self_attention.query_key_value.weight", "transformer.h.4.self_attention.query_key_value.bias", "transformer.h.4.self_attention.dense.weight", "transformer.h.4.self_attention.dense.bias", "transformer.h.4.post_attention_layernorm.weight", "transformer.h.4.post_attention_layernorm.bias", "transformer.h.4.mlp.dense_h_to_4h.weight", "transformer.h.4.mlp.dense_h_to_4h.bias", "transformer.h.4.mlp.dense_4h_to_h.weight", "transformer.h.4.mlp.dense_4h_to_h.bias", "transformer.h.5.input_layernorm.weight", "transformer.h.5.input_layernorm.bias", "transformer.h.5.self_attention.query_key_value.weight", "transformer.h.5.self_attention.query_key_value.bias", "transformer.h.5.self_attention.dense.weight", "transformer.h.5.self_attention.dense.bias", "transformer.h.5.post_attention_layernorm.weight", "transformer.h.5.post_attention_layernorm.bias", "transformer.h.5.mlp.dense_h_to_4h.weight", "transformer.h.5.mlp.dense_h_to_4h.bias", "transformer.h.5.mlp.dense_4h_to_h.weight", "transformer.h.5.mlp.dense_4h_to_h.bias", "transformer.h.6.input_layernorm.weight", "transformer.h.6.input_layernorm.bias", "transformer.h.6.self_attention.query_key_value.weight", "transformer.h.6.self_attention.query_key_value.bias", "transformer.h.6.self_attention.dense.weight", "transformer.h.6.self_attention.dense.bias", "transformer.h.6.post_attention_layernorm.weight", "transformer.h.6.post_attention_layernorm.bias", "transformer.h.6.mlp.dense_h_to_4h.weight", "transformer.h.6.mlp.dense_h_to_4h.bias", "transformer.h.6.mlp.dense_4h_to_h.weight", "transformer.h.6.mlp.dense_4h_to_h.bias", "transformer.h.7.input_layernorm.weight", "transformer.h.7.input_layernorm.bias", "transformer.h.7.self_attention.query_key_value.weight", "transformer.h.7.self_attention.query_key_value.bias", "transformer.h.7.self_attention.dense.weight", "transformer.h.7.self_attention.dense.bias", "transformer.h.7.post_attention_layernorm.weight", "transformer.h.7.post_attention_layernorm.bias", "transformer.h.7.mlp.dense_h_to_4h.weight", "transformer.h.7.mlp.dense_h_to_4h.bias", "transformer.h.7.mlp.dense_4h_to_h.weight", "transformer.h.7.mlp.dense_4h_to_h.bias", "transformer.h.8.input_layernorm.weight", "transformer.h.8.input_layernorm.bias", "transformer.h.8.self_attention.query_key_value.weight", "transformer.h.8.self_attention.query_key_value.bias", "transformer.h.8.self_attention.dense.weight", "transformer.h.8.self_attention.dense.bias", "transformer.h.8.post_attention_layernorm.weight", "transformer.h.8.post_attention_layernorm.bias", "transformer.h.8.mlp.dense_h_to_4h.weight", "transformer.h.8.mlp.dense_h_to_4h.bias", "transformer.h.8.mlp.dense_4h_to_h.weight", "transformer.h.8.mlp.dense_4h_to_h.bias", "transformer.h.9.input_layernorm.weight", "transformer.h.9.input_layernorm.bias", "transformer.h.9.self_attention.query_key_value.weight", "transformer.h.9.self_attention.query_key_value.bias", "transformer.h.9.self_attention.dense.weight", "transformer.h.9.self_attention.dense.bias", "transformer.h.9.post_attention_layernorm.weight", "transformer.h.9.post_attention_layernorm.bias", "transformer.h.9.mlp.dense_h_to_4h.weight", "transformer.h.9.mlp.dense_h_to_4h.bias", "transformer.h.9.mlp.dense_4h_to_h.weight", "transformer.h.9.mlp.dense_4h_to_h.bias", "transformer.h.10.input_layernorm.weight", "transformer.h.10.input_layernorm.bias", "transformer.h.10.self_attention.query_key_value.weight", "transformer.h.10.self_attention.query_key_value.bias", "transformer.h.10.self_attention.dense.weight", "transformer.h.10.self_attention.dense.bias", "transformer.h.10.post_attention_layernorm.weight", "transformer.h.10.post_attention_layernorm.bias", "transformer.h.10.mlp.dense_h_to_4h.weight", "transformer.h.10.mlp.dense_h_to_4h.bias", "transformer.h.10.mlp.dense_4h_to_h.weight", "transformer.h.10.mlp.dense_4h_to_h.bias", "transformer.h.11.input_layernorm.weight", "transformer.h.11.input_layernorm.bias", "transformer.h.11.self_attention.query_key_value.weight", "transformer.h.11.self_attention.query_key_value.bias", "transformer.h.11.self_attention.dense.weight", "transformer.h.11.self_attention.dense.bias", "transformer.h.11.post_attention_layernorm.weight", "transformer.h.11.post_attention_layernorm.bias", "transformer.h.11.mlp.dense_h_to_4h.weight", "transformer.h.11.mlp.dense_h_to_4h.bias", "transformer.h.11.mlp.dense_4h_to_h.weight", "transformer.h.11.mlp.dense_4h_to_h.bias", "transformer.h.12.input_layernorm.weight", "transformer.h.12.input_layernorm.bias", "transformer.h.12.self_attention.query_key_value.weight", "transformer.h.12.self_attention.query_key_value.bias", "transformer.h.12.self_attention.dense.weight", "transformer.h.12.self_attention.dense.bias", "transformer.h.12.post_attention_layernorm.weight", "transformer.h.12.post_attention_layernorm.bias", "transformer.h.12.mlp.dense_h_to_4h.weight", "transformer.h.12.mlp.dense_h_to_4h.bias", "transformer.h.12.mlp.dense_4h_to_h.weight", "transformer.h.12.mlp.dense_4h_to_h.bias", "transformer.h.13.input_layernorm.weight", "transformer.h.13.input_layernorm.bias", "transformer.h.13.self_attention.query_key_value.weight", "transformer.h.13.self_attention.query_key_value.bias", "transformer.h.13.self_attention.dense.weight", "transformer.h.13.self_attention.dense.bias", "transformer.h.13.post_attention_layernorm.weight", "transformer.h.13.post_attention_layernorm.bias", "transformer.h.13.mlp.dense_h_to_4h.weight", "transformer.h.13.mlp.dense_h_to_4h.bias", "transformer.h.13.mlp.dense_4h_to_h.weight", "transformer.h.13.mlp.dense_4h_to_h.bias", "transformer.h.14.input_layernorm.weight", "transformer.h.14.input_layernorm.bias", "transformer.h.14.self_attention.query_key_value.weight", "transformer.h.14.self_attention.query_key_value.bias", "transformer.h.14.self_attention.dense.weight", "transformer.h.14.self_attention.dense.bias", "transformer.h.14.post_attention_layernorm.weight", "transformer.h.14.post_attention_layernorm.bias", "transformer.h.14.mlp.dense_h_to_4h.weight", "transformer.h.14.mlp.dense_h_to_4h.bias", "transformer.h.14.mlp.dense_4h_to_h.weight", "transformer.h.14.mlp.dense_4h_to_h.bias", "transformer.h.15.input_layernorm.weight", "transformer.h.15.input_layernorm.bias", "transformer.h.15.self_attention.query_key_value.weight", "transformer.h.15.self_attention.query_key_value.bias", "transformer.h.15.self_attention.dense.weight", "transformer.h.15.self_attention.dense.bias", "transformer.h.15.post_attention_layernorm.weight", "transformer.h.15.post_attention_layernorm.bias", "transformer.h.15.mlp.dense_h_to_4h.weight", "transformer.h.15.mlp.dense_h_to_4h.bias", "transformer.h.15.mlp.dense_4h_to_h.weight", "transformer.h.15.mlp.dense_4h_to_h.bias", "transformer.h.16.input_layernorm.weight", "transformer.h.16.input_layernorm.bias", "transformer.h.16.self_attention.query_key_value.weight", "transformer.h.16.self_attention.query_key_value.bias", "transformer.h.16.self_attention.dense.weight", "transformer.h.16.self_attention.dense.bias", "transformer.h.16.post_attention_layernorm.weight", "transformer.h.16.post_attention_layernorm.bias", "transformer.h.16.mlp.dense_h_to_4h.weight", "transformer.h.16.mlp.dense_h_to_4h.bias", "transformer.h.16.mlp.dense_4h_to_h.weight", "transformer.h.16.mlp.dense_4h_to_h.bias", "transformer.h.17.input_layernorm.weight", "transformer.h.17.input_layernorm.bias", "transformer.h.17.self_attention.query_key_value.weight", "transformer.h.17.self_attention.query_key_value.bias", "transformer.h.17.self_attention.dense.weight", "transformer.h.17.self_attention.dense.bias", "transformer.h.17.post_attention_layernorm.weight", "transformer.h.17.post_attention_layernorm.bias", "transformer.h.17.mlp.dense_h_to_4h.weight", "transformer.h.17.mlp.dense_h_to_4h.bias", "transformer.h.17.mlp.dense_4h_to_h.weight", "transformer.h.17.mlp.dense_4h_to_h.bias", "transformer.h.18.input_layernorm.weight", "transformer.h.18.input_layernorm.bias", "transformer.h.18.self_attention.query_key_value.weight", "transformer.h.18.self_attention.query_key_value.bias", "transformer.h.18.self_attention.dense.weight", "transformer.h.18.self_attention.dense.bias", "transformer.h.18.post_attention_layernorm.weight", "transformer.h.18.post_attention_layernorm.bias", "transformer.h.18.mlp.dense_h_to_4h.weight", "transformer.h.18.mlp.dense_h_to_4h.bias", "transformer.h.18.mlp.dense_4h_to_h.weight", "transformer.h.18.mlp.dense_4h_to_h.bias", "transformer.h.19.input_layernorm.weight", "transformer.h.19.input_layernorm.bias", "transformer.h.19.self_attention.query_key_value.weight", "transformer.h.19.self_attention.query_key_value.bias", "transformer.h.19.self_attention.dense.weight", "transformer.h.19.self_attention.dense.bias", "transformer.h.19.post_attention_layernorm.weight", "transformer.h.19.post_attention_layernorm.bias", "transformer.h.19.mlp.dense_h_to_4h.weight", "transformer.h.19.mlp.dense_h_to_4h.bias", "transformer.h.19.mlp.dense_4h_to_h.weight", "transformer.h.19.mlp.dense_4h_to_h.bias", "transformer.h.20.input_layernorm.weight", "transformer.h.20.input_layernorm.bias", "transformer.h.20.self_attention.query_key_value.weight", "transformer.h.20.self_attention.query_key_value.bias", "transformer.h.20.self_attention.dense.weight", "transformer.h.20.self_attention.dense.bias", "transformer.h.20.post_attention_layernorm.weight", "transformer.h.20.post_attention_layernorm.bias", "transformer.h.20.mlp.dense_h_to_4h.weight", "transformer.h.20.mlp.dense_h_to_4h.bias", "transformer.h.20.mlp.dense_4h_to_h.weight", "transformer.h.20.mlp.dense_4h_to_h.bias", "transformer.h.21.input_layernorm.weight", "transformer.h.21.input_layernorm.bias", "transformer.h.21.self_attention.query_key_value.weight", "transformer.h.21.self_attention.query_key_value.bias", "transformer.h.21.self_attention.dense.weight", "transformer.h.21.self_attention.dense.bias", "transformer.h.21.post_attention_layernorm.weight", "transformer.h.21.post_attention_layernorm.bias", "transformer.h.21.mlp.dense_h_to_4h.weight", "transformer.h.21.mlp.dense_h_to_4h.bias", "transformer.h.21.mlp.dense_4h_to_h.weight", "transformer.h.21.mlp.dense_4h_to_h.bias", "transformer.h.22.input_layernorm.weight", "transformer.h.22.input_layernorm.bias", "transformer.h.22.self_attention.query_key_value.weight", "transformer.h.22.self_attention.query_key_value.bias", "transformer.h.22.self_attention.dense.weight", "transformer.h.22.self_attention.dense.bias", "transformer.h.22.post_attention_layernorm.weight", "transformer.h.22.post_attention_layernorm.bias", "transformer.h.22.mlp.dense_h_to_4h.weight", "transformer.h.22.mlp.dense_h_to_4h.bias", "transformer.h.22.mlp.dense_4h_to_h.weight", "transformer.h.22.mlp.dense_4h_to_h.bias", "transformer.h.23.input_layernorm.weight", "transformer.h.23.input_layernorm.bias", "transformer.h.23.self_attention.query_key_value.weight", "transformer.h.23.self_attention.query_key_value.bias", "transformer.h.23.self_attention.dense.weight", "transformer.h.23.self_attention.dense.bias", "transformer.h.23.post_attention_layernorm.weight", "transformer.h.23.post_attention_layernorm.bias", "transformer.h.23.mlp.dense_h_to_4h.weight", "transformer.h.23.mlp.dense_h_to_4h.bias", "transformer.h.23.mlp.dense_4h_to_h.weight", "transformer.h.23.mlp.dense_4h_to_h.bias", "transformer.ln_f.weight", "transformer.ln_f.bias", "lm_head.weight"]}