A model from - https://huggingface.co/aisingapore/sea-lion-7b-instruct-gguf/blob/main/sea-lion-7b-instruct-Q4_K_M.gguf
8 Pulls Updated 6 months ago
af0ebb4645be · 5.0GB
-
general.architecturempt
-
general.file_type15
-
general.namesea-lion-7b-instruct
-
general.quantization_version2
-
mpt.attention.head_count32
-
mpt.attention.layer_norm_epsilon1e-05
-
mpt.attention.max_alibi_bias0
-
mpt.block_count32
-
mpt.context_length2048
-
mpt.embedding_length4096
-
mpt.feed_forward_length16384
-
tokenizer.ggml.add_bos_tokenfalse
-
tokenizer.ggml.add_eos_tokenfalse
-
tokenizer.ggml.eos_token_id1
-
tokenizer.ggml.modelllama
-
tokenizer.ggml.padding_token_id3
-
tokenizer.ggml.scores[0, 0, 0, 0, 0, ...]
-
tokenizer.ggml.token_type[2, 3, 3, 3, 1, ...]
-
tokenizer.ggml.tokens[<unk>, <|endoftext|>, <|endofline|>, <|padding|>, , ...]
-
tokenizer.ggml.unknown_token_id0
-
NameTypeShape
-
token_embd.weightQ6_K[4096, 256000]
-
blk.0.attn_k_norm.biasF32[4096]
-
blk.0.attn_k_norm.weightF32[4096]
-
blk.0.attn_norm.biasF32[4096]
-
blk.0.attn_norm.weightF32[4096]
-
blk.0.attn_output.biasF32[4096]
-
blk.0.attn_output.weightQ4_K[4096, 4096]
-
blk.0.attn_q_norm.biasF32[4096]
-
blk.0.attn_q_norm.weightF32[4096]
-
blk.0.attn_qkv.biasF32[12288]
-
blk.0.attn_qkv.weightQ5_K[4096, 12288]
-
blk.0.ffn_down.biasF32[4096]
-
blk.0.ffn_down.weightQ6_K[16384, 4096]
-
blk.0.ffn_norm.biasF32[4096]
-
blk.0.ffn_norm.weightF32[4096]
-
blk.0.ffn_up.biasF32[16384]
-
blk.0.ffn_up.weightQ4_K[4096, 16384]
-
blk.1.attn_k_norm.biasF32[4096]
-
blk.1.attn_k_norm.weightF32[4096]
-
blk.1.attn_norm.biasF32[4096]
-
blk.1.attn_norm.weightF32[4096]
-
blk.1.attn_output.biasF32[4096]
-
blk.1.attn_output.weightQ4_K[4096, 4096]
-
blk.1.attn_q_norm.biasF32[4096]
-
blk.1.attn_q_norm.weightF32[4096]
-
blk.1.attn_qkv.biasF32[12288]
-
blk.1.attn_qkv.weightQ5_K[4096, 12288]
-
blk.1.ffn_down.biasF32[4096]
-
blk.1.ffn_down.weightQ6_K[16384, 4096]
-
blk.1.ffn_norm.biasF32[4096]
-
blk.1.ffn_norm.weightF32[4096]
-
blk.1.ffn_up.biasF32[16384]
-
blk.1.ffn_up.weightQ4_K[4096, 16384]
-
blk.2.attn_k_norm.biasF32[4096]
-
blk.2.attn_k_norm.weightF32[4096]
-
blk.2.attn_norm.biasF32[4096]
-
blk.2.attn_norm.weightF32[4096]
-
blk.2.attn_output.biasF32[4096]
-
blk.2.attn_output.weightQ4_K[4096, 4096]
-
blk.2.attn_q_norm.biasF32[4096]
-
blk.2.attn_q_norm.weightF32[4096]
-
blk.2.attn_qkv.biasF32[12288]
-
blk.2.attn_qkv.weightQ5_K[4096, 12288]
-
blk.2.ffn_down.biasF32[4096]
-
blk.2.ffn_down.weightQ6_K[16384, 4096]
-
blk.2.ffn_norm.biasF32[4096]
-
blk.2.ffn_norm.weightF32[4096]
-
blk.2.ffn_up.biasF32[16384]
-
blk.2.ffn_up.weightQ4_K[4096, 16384]
-
blk.3.attn_k_norm.biasF32[4096]
-
blk.3.attn_k_norm.weightF32[4096]
-
blk.3.attn_norm.biasF32[4096]
-
blk.3.attn_norm.weightF32[4096]
-
blk.3.attn_output.biasF32[4096]
-
blk.3.attn_output.weightQ4_K[4096, 4096]
-
blk.3.attn_q_norm.biasF32[4096]
-
blk.3.attn_q_norm.weightF32[4096]
-
blk.3.attn_qkv.biasF32[12288]
-
blk.3.attn_qkv.weightQ5_K[4096, 12288]
-
blk.3.ffn_down.biasF32[4096]
-
blk.3.ffn_down.weightQ6_K[16384, 4096]
-
blk.3.ffn_norm.biasF32[4096]
-
blk.3.ffn_norm.weightF32[4096]
-
blk.3.ffn_up.biasF32[16384]
-
blk.3.ffn_up.weightQ4_K[4096, 16384]
-
blk.4.attn_k_norm.biasF32[4096]
-
blk.4.attn_k_norm.weightF32[4096]
-
blk.4.attn_norm.biasF32[4096]
-
blk.4.attn_norm.weightF32[4096]
-
blk.4.attn_output.biasF32[4096]
-
blk.4.attn_output.weightQ4_K[4096, 4096]
-
blk.4.attn_q_norm.biasF32[4096]
-
blk.4.attn_q_norm.weightF32[4096]
-
blk.4.attn_qkv.biasF32[12288]
-
blk.4.attn_qkv.weightQ5_K[4096, 12288]
-
blk.4.ffn_down.biasF32[4096]
-
blk.4.ffn_down.weightQ4_K[16384, 4096]
-
blk.4.ffn_norm.biasF32[4096]
-
blk.4.ffn_norm.weightF32[4096]
-
blk.4.ffn_up.biasF32[16384]
-
blk.4.ffn_up.weightQ4_K[4096, 16384]
-
blk.5.attn_k_norm.biasF32[4096]
-
blk.5.attn_k_norm.weightF32[4096]
-
blk.5.attn_norm.biasF32[4096]
-
blk.5.attn_norm.weightF32[4096]
-
blk.5.attn_output.biasF32[4096]
-
blk.5.attn_output.weightQ4_K[4096, 4096]
-
blk.5.attn_q_norm.biasF32[4096]
-
blk.5.attn_q_norm.weightF32[4096]
-
blk.5.attn_qkv.biasF32[12288]
-
blk.5.attn_qkv.weightQ5_K[4096, 12288]
-
blk.5.ffn_down.biasF32[4096]
-
blk.5.ffn_down.weightQ4_K[16384, 4096]
-
blk.5.ffn_norm.biasF32[4096]
-
blk.5.ffn_norm.weightF32[4096]
-
blk.5.ffn_up.biasF32[16384]
-
blk.5.ffn_up.weightQ4_K[4096, 16384]
-
blk.6.attn_k_norm.biasF32[4096]
-
blk.6.attn_k_norm.weightF32[4096]
-
blk.6.attn_norm.biasF32[4096]
-
blk.6.attn_norm.weightF32[4096]
-
blk.6.attn_output.biasF32[4096]
-
blk.6.attn_output.weightQ4_K[4096, 4096]
-
blk.6.attn_q_norm.biasF32[4096]
-
blk.6.attn_q_norm.weightF32[4096]
-
blk.6.attn_qkv.biasF32[12288]
-
blk.6.attn_qkv.weightQ5_K[4096, 12288]
-
blk.6.ffn_down.biasF32[4096]
-
blk.6.ffn_down.weightQ6_K[16384, 4096]
-
blk.6.ffn_norm.biasF32[4096]
-
blk.6.ffn_norm.weightF32[4096]
-
blk.6.ffn_up.biasF32[16384]
-
blk.6.ffn_up.weightQ4_K[4096, 16384]
-
blk.7.attn_k_norm.biasF32[4096]
-
blk.7.attn_k_norm.weightF32[4096]
-
blk.7.attn_norm.biasF32[4096]
-
blk.7.attn_norm.weightF32[4096]
-
blk.7.attn_output.biasF32[4096]
-
blk.7.attn_output.weightQ4_K[4096, 4096]
-
blk.7.attn_q_norm.biasF32[4096]
-
blk.7.attn_q_norm.weightF32[4096]
-
blk.7.attn_qkv.biasF32[12288]
-
blk.7.attn_qkv.weightQ5_K[4096, 12288]
-
blk.7.ffn_down.biasF32[4096]
-
blk.7.ffn_down.weightQ4_K[16384, 4096]
-
blk.7.ffn_norm.biasF32[4096]
-
blk.7.ffn_norm.weightF32[4096]
-
blk.7.ffn_up.biasF32[16384]
-
blk.7.ffn_up.weightQ4_K[4096, 16384]
-
blk.8.attn_k_norm.biasF32[4096]
-
blk.8.attn_k_norm.weightF32[4096]
-
blk.8.attn_norm.biasF32[4096]
-
blk.8.attn_norm.weightF32[4096]
-
blk.8.attn_output.biasF32[4096]
-
blk.8.attn_output.weightQ4_K[4096, 4096]
-
blk.8.attn_q_norm.biasF32[4096]
-
blk.8.attn_q_norm.weightF32[4096]
-
blk.8.attn_qkv.biasF32[12288]
-
blk.8.attn_qkv.weightQ5_K[4096, 12288]
-
blk.8.ffn_down.biasF32[4096]
-
blk.8.ffn_down.weightQ4_K[16384, 4096]
-
blk.8.ffn_norm.biasF32[4096]
-
blk.8.ffn_norm.weightF32[4096]
-
blk.8.ffn_up.biasF32[16384]
-
blk.8.ffn_up.weightQ4_K[4096, 16384]
-
blk.9.attn_k_norm.biasF32[4096]
-
blk.9.attn_k_norm.weightF32[4096]
-
blk.9.attn_norm.biasF32[4096]
-
blk.9.attn_norm.weightF32[4096]
-
blk.9.attn_output.biasF32[4096]
-
blk.9.attn_output.weightQ4_K[4096, 4096]
-
blk.9.attn_q_norm.biasF32[4096]
-
blk.9.attn_q_norm.weightF32[4096]
-
blk.9.attn_qkv.biasF32[12288]
-
blk.9.attn_qkv.weightQ5_K[4096, 12288]
-
blk.9.ffn_down.biasF32[4096]
-
blk.9.ffn_down.weightQ6_K[16384, 4096]
-
blk.9.ffn_norm.biasF32[4096]
-
blk.9.ffn_norm.weightF32[4096]
-
blk.9.ffn_up.biasF32[16384]
-
blk.9.ffn_up.weightQ4_K[4096, 16384]
-
blk.10.attn_k_norm.biasF32[4096]
-
blk.10.attn_k_norm.weightF32[4096]
-
blk.10.attn_norm.biasF32[4096]
-
blk.10.attn_norm.weightF32[4096]
-
blk.10.attn_output.biasF32[4096]
-
blk.10.attn_output.weightQ4_K[4096, 4096]
-
blk.10.attn_q_norm.biasF32[4096]
-
blk.10.attn_q_norm.weightF32[4096]
-
blk.10.attn_qkv.biasF32[12288]
-
blk.10.attn_qkv.weightQ5_K[4096, 12288]
-
blk.10.ffn_down.biasF32[4096]
-
blk.10.ffn_down.weightQ4_K[16384, 4096]
-
blk.10.ffn_norm.biasF32[4096]
-
blk.10.ffn_norm.weightF32[4096]
-
blk.10.ffn_up.biasF32[16384]
-
blk.10.ffn_up.weightQ4_K[4096, 16384]
-
blk.11.attn_k_norm.biasF32[4096]
-
blk.11.attn_k_norm.weightF32[4096]
-
blk.11.attn_norm.biasF32[4096]
-
blk.11.attn_norm.weightF32[4096]
-
blk.11.attn_output.biasF32[4096]
-
blk.11.attn_output.weightQ4_K[4096, 4096]
-
blk.11.attn_q_norm.biasF32[4096]
-
blk.11.attn_q_norm.weightF32[4096]
-
blk.11.attn_qkv.biasF32[12288]
-
blk.11.attn_qkv.weightQ5_K[4096, 12288]
-
blk.11.ffn_down.biasF32[4096]
-
blk.11.ffn_down.weightQ4_K[16384, 4096]
-
blk.11.ffn_norm.biasF32[4096]
-
blk.11.ffn_norm.weightF32[4096]
-
blk.11.ffn_up.biasF32[16384]
-
blk.11.ffn_up.weightQ4_K[4096, 16384]
-
blk.12.attn_k_norm.biasF32[4096]
-
blk.12.attn_k_norm.weightF32[4096]
-
blk.12.attn_norm.biasF32[4096]
-
blk.12.attn_norm.weightF32[4096]
-
blk.12.attn_output.biasF32[4096]
-
blk.12.attn_output.weightQ4_K[4096, 4096]
-
blk.12.attn_q_norm.biasF32[4096]
-
blk.12.attn_q_norm.weightF32[4096]
-
blk.12.attn_qkv.biasF32[12288]
-
blk.12.attn_qkv.weightQ5_K[4096, 12288]
-
blk.12.ffn_down.biasF32[4096]
-
blk.12.ffn_down.weightQ6_K[16384, 4096]
-
blk.12.ffn_norm.biasF32[4096]
-
blk.12.ffn_norm.weightF32[4096]
-
blk.12.ffn_up.biasF32[16384]
-
blk.12.ffn_up.weightQ4_K[4096, 16384]
-
blk.13.attn_k_norm.biasF32[4096]
-
blk.13.attn_k_norm.weightF32[4096]
-
blk.13.attn_norm.biasF32[4096]
-
blk.13.attn_norm.weightF32[4096]
-
blk.13.attn_output.biasF32[4096]
-
blk.13.attn_output.weightQ4_K[4096, 4096]
-
blk.13.attn_q_norm.biasF32[4096]
-
blk.13.attn_q_norm.weightF32[4096]
-
blk.13.attn_qkv.biasF32[12288]
-
blk.13.attn_qkv.weightQ5_K[4096, 12288]
-
blk.13.ffn_down.biasF32[4096]
-
blk.13.ffn_down.weightQ4_K[16384, 4096]
-
blk.13.ffn_norm.biasF32[4096]
-
blk.13.ffn_norm.weightF32[4096]
-
blk.13.ffn_up.biasF32[16384]
-
blk.13.ffn_up.weightQ4_K[4096, 16384]
-
blk.14.attn_k_norm.biasF32[4096]
-
blk.14.attn_k_norm.weightF32[4096]
-
blk.14.attn_norm.biasF32[4096]
-
blk.14.attn_norm.weightF32[4096]
-
blk.14.attn_output.biasF32[4096]
-
blk.14.attn_output.weightQ4_K[4096, 4096]
-
blk.14.attn_q_norm.biasF32[4096]
-
blk.14.attn_q_norm.weightF32[4096]
-
blk.14.attn_qkv.biasF32[12288]
-
blk.14.attn_qkv.weightQ5_K[4096, 12288]
-
blk.14.ffn_down.biasF32[4096]
-
blk.14.ffn_down.weightQ4_K[16384, 4096]
-
blk.14.ffn_norm.biasF32[4096]
-
blk.14.ffn_norm.weightF32[4096]
-
blk.14.ffn_up.biasF32[16384]
-
blk.14.ffn_up.weightQ4_K[4096, 16384]
-
blk.15.attn_k_norm.biasF32[4096]
-
blk.15.attn_k_norm.weightF32[4096]
-
blk.15.attn_norm.biasF32[4096]
-
blk.15.attn_norm.weightF32[4096]
-
blk.15.attn_output.biasF32[4096]
-
blk.15.attn_output.weightQ4_K[4096, 4096]
-
blk.15.attn_q_norm.biasF32[4096]
-
blk.15.attn_q_norm.weightF32[4096]
-
blk.15.attn_qkv.biasF32[12288]
-
blk.15.attn_qkv.weightQ5_K[4096, 12288]
-
blk.15.ffn_down.biasF32[4096]
-
blk.15.ffn_down.weightQ6_K[16384, 4096]
-
blk.15.ffn_norm.biasF32[4096]
-
blk.15.ffn_norm.weightF32[4096]
-
blk.15.ffn_up.biasF32[16384]
-
blk.15.ffn_up.weightQ4_K[4096, 16384]
-
blk.16.attn_k_norm.biasF32[4096]
-
blk.16.attn_k_norm.weightF32[4096]
-
blk.16.attn_norm.biasF32[4096]
-
blk.16.attn_norm.weightF32[4096]
-
blk.16.attn_output.biasF32[4096]
-
blk.16.attn_output.weightQ4_K[4096, 4096]
-
blk.16.attn_q_norm.biasF32[4096]
-
blk.16.attn_q_norm.weightF32[4096]
-
blk.16.attn_qkv.biasF32[12288]
-
blk.16.attn_qkv.weightQ5_K[4096, 12288]
-
blk.16.ffn_down.biasF32[4096]
-
blk.16.ffn_down.weightQ4_K[16384, 4096]
-
blk.16.ffn_norm.biasF32[4096]
-
blk.16.ffn_norm.weightF32[4096]
-
blk.16.ffn_up.biasF32[16384]
-
blk.16.ffn_up.weightQ4_K[4096, 16384]
-
blk.17.attn_k_norm.biasF32[4096]
-
blk.17.attn_k_norm.weightF32[4096]
-
blk.17.attn_norm.biasF32[4096]
-
blk.17.attn_norm.weightF32[4096]
-
blk.17.attn_output.biasF32[4096]
-
blk.17.attn_output.weightQ4_K[4096, 4096]
-
blk.17.attn_q_norm.biasF32[4096]
-
blk.17.attn_q_norm.weightF32[4096]
-
blk.17.attn_qkv.biasF32[12288]
-
blk.17.attn_qkv.weightQ5_K[4096, 12288]
-
blk.17.ffn_down.biasF32[4096]
-
blk.17.ffn_down.weightQ4_K[16384, 4096]
-
blk.17.ffn_norm.biasF32[4096]
-
blk.17.ffn_norm.weightF32[4096]
-
blk.17.ffn_up.biasF32[16384]
-
blk.17.ffn_up.weightQ4_K[4096, 16384]
-
blk.18.attn_k_norm.biasF32[4096]
-
blk.18.attn_k_norm.weightF32[4096]
-
blk.18.attn_norm.biasF32[4096]
-
blk.18.attn_norm.weightF32[4096]
-
blk.18.attn_output.biasF32[4096]
-
blk.18.attn_output.weightQ4_K[4096, 4096]
-
blk.18.attn_q_norm.biasF32[4096]
-
blk.18.attn_q_norm.weightF32[4096]
-
blk.18.attn_qkv.biasF32[12288]
-
blk.18.attn_qkv.weightQ5_K[4096, 12288]
-
blk.18.ffn_down.biasF32[4096]
-
blk.18.ffn_down.weightQ6_K[16384, 4096]
-
blk.18.ffn_norm.biasF32[4096]
-
blk.18.ffn_norm.weightF32[4096]
-
blk.18.ffn_up.biasF32[16384]
-
blk.18.ffn_up.weightQ4_K[4096, 16384]
-
blk.19.attn_k_norm.biasF32[4096]
-
blk.19.attn_k_norm.weightF32[4096]
-
blk.19.attn_norm.biasF32[4096]
-
blk.19.attn_norm.weightF32[4096]
-
blk.19.attn_output.biasF32[4096]
-
blk.19.attn_output.weightQ4_K[4096, 4096]
-
blk.19.attn_q_norm.biasF32[4096]
-
blk.19.attn_q_norm.weightF32[4096]
-
blk.19.attn_qkv.biasF32[12288]
-
blk.19.attn_qkv.weightQ5_K[4096, 12288]
-
blk.19.ffn_down.biasF32[4096]
-
blk.19.ffn_down.weightQ4_K[16384, 4096]
-
blk.19.ffn_norm.biasF32[4096]
-
blk.19.ffn_norm.weightF32[4096]
-
blk.19.ffn_up.biasF32[16384]
-
blk.19.ffn_up.weightQ4_K[4096, 16384]
-
blk.20.attn_k_norm.biasF32[4096]
-
blk.20.attn_k_norm.weightF32[4096]
-
blk.20.attn_norm.biasF32[4096]
-
blk.20.attn_norm.weightF32[4096]
-
blk.20.attn_output.biasF32[4096]
-
blk.20.attn_output.weightQ4_K[4096, 4096]
-
blk.20.attn_q_norm.biasF32[4096]
-
blk.20.attn_q_norm.weightF32[4096]
-
blk.20.attn_qkv.biasF32[12288]
-
blk.20.attn_qkv.weightQ5_K[4096, 12288]
-
blk.20.ffn_down.biasF32[4096]
-
blk.20.ffn_down.weightQ4_K[16384, 4096]
-
blk.20.ffn_norm.biasF32[4096]
-
blk.20.ffn_norm.weightF32[4096]
-
blk.20.ffn_up.biasF32[16384]
-
blk.20.ffn_up.weightQ4_K[4096, 16384]
-
blk.21.attn_k_norm.biasF32[4096]
-
blk.21.attn_k_norm.weightF32[4096]
-
blk.21.attn_norm.biasF32[4096]
-
blk.21.attn_norm.weightF32[4096]
-
blk.21.attn_output.biasF32[4096]
-
blk.21.attn_output.weightQ4_K[4096, 4096]
-
blk.21.attn_q_norm.biasF32[4096]
-
blk.21.attn_q_norm.weightF32[4096]
-
blk.21.attn_qkv.biasF32[12288]
-
blk.21.attn_qkv.weightQ5_K[4096, 12288]
-
blk.21.ffn_down.biasF32[4096]
-
blk.21.ffn_down.weightQ6_K[16384, 4096]
-
blk.21.ffn_norm.biasF32[4096]
-
blk.21.ffn_norm.weightF32[4096]
-
blk.21.ffn_up.biasF32[16384]
-
blk.21.ffn_up.weightQ4_K[4096, 16384]
-
blk.22.attn_k_norm.biasF32[4096]
-
blk.22.attn_k_norm.weightF32[4096]
-
blk.22.attn_norm.biasF32[4096]
-
blk.22.attn_norm.weightF32[4096]
-
blk.22.attn_output.biasF32[4096]
-
blk.22.attn_output.weightQ4_K[4096, 4096]
-
blk.22.attn_q_norm.biasF32[4096]
-
blk.22.attn_q_norm.weightF32[4096]
-
blk.22.attn_qkv.biasF32[12288]
-
blk.22.attn_qkv.weightQ5_K[4096, 12288]
-
blk.22.ffn_down.biasF32[4096]
-
blk.22.ffn_down.weightQ4_K[16384, 4096]
-
blk.22.ffn_norm.biasF32[4096]
-
blk.22.ffn_norm.weightF32[4096]
-
blk.22.ffn_up.biasF32[16384]
-
blk.22.ffn_up.weightQ4_K[4096, 16384]
-
blk.23.attn_k_norm.biasF32[4096]
-
blk.23.attn_k_norm.weightF32[4096]
-
blk.23.attn_norm.biasF32[4096]
-
blk.23.attn_norm.weightF32[4096]
-
blk.23.attn_output.biasF32[4096]
-
blk.23.attn_output.weightQ4_K[4096, 4096]
-
blk.23.attn_q_norm.biasF32[4096]
-
blk.23.attn_q_norm.weightF32[4096]
-
blk.23.attn_qkv.biasF32[12288]
-
blk.23.attn_qkv.weightQ5_K[4096, 12288]
-
blk.23.ffn_down.biasF32[4096]
-
blk.23.ffn_down.weightQ4_K[16384, 4096]
-
blk.23.ffn_norm.biasF32[4096]
-
blk.23.ffn_norm.weightF32[4096]
-
blk.23.ffn_up.biasF32[16384]
-
blk.23.ffn_up.weightQ4_K[4096, 16384]
-
blk.24.attn_k_norm.biasF32[4096]
-
blk.24.attn_k_norm.weightF32[4096]
-
blk.24.attn_norm.biasF32[4096]
-
blk.24.attn_norm.weightF32[4096]
-
blk.24.attn_output.biasF32[4096]
-
blk.24.attn_output.weightQ4_K[4096, 4096]
-
blk.24.attn_q_norm.biasF32[4096]
-
blk.24.attn_q_norm.weightF32[4096]
-
blk.24.attn_qkv.biasF32[12288]
-
blk.24.attn_qkv.weightQ5_K[4096, 12288]
-
blk.24.ffn_down.biasF32[4096]
-
blk.24.ffn_down.weightQ6_K[16384, 4096]
-
blk.24.ffn_norm.biasF32[4096]
-
blk.24.ffn_norm.weightF32[4096]
-
blk.24.ffn_up.biasF32[16384]
-
blk.24.ffn_up.weightQ4_K[4096, 16384]
-
blk.25.attn_k_norm.biasF32[4096]
-
blk.25.attn_k_norm.weightF32[4096]
-
blk.25.attn_norm.biasF32[4096]
-
blk.25.attn_norm.weightF32[4096]
-
blk.25.attn_output.biasF32[4096]
-
blk.25.attn_output.weightQ4_K[4096, 4096]
-
blk.25.attn_q_norm.biasF32[4096]
-
blk.25.attn_q_norm.weightF32[4096]
-
blk.25.attn_qkv.biasF32[12288]
-
blk.25.attn_qkv.weightQ5_K[4096, 12288]
-
blk.25.ffn_down.biasF32[4096]
-
blk.25.ffn_down.weightQ4_K[16384, 4096]
-
blk.25.ffn_norm.biasF32[4096]
-
blk.25.ffn_norm.weightF32[4096]
-
blk.25.ffn_up.biasF32[16384]
-
blk.25.ffn_up.weightQ4_K[4096, 16384]
-
blk.26.attn_k_norm.biasF32[4096]
-
blk.26.attn_k_norm.weightF32[4096]
-
blk.26.attn_norm.biasF32[4096]
-
blk.26.attn_norm.weightF32[4096]
-
blk.26.attn_output.biasF32[4096]
-
blk.26.attn_output.weightQ4_K[4096, 4096]
-
blk.26.attn_q_norm.biasF32[4096]
-
blk.26.attn_q_norm.weightF32[4096]
-
blk.26.attn_qkv.biasF32[12288]
-
blk.26.attn_qkv.weightQ5_K[4096, 12288]
-
blk.26.ffn_down.biasF32[4096]
-
blk.26.ffn_down.weightQ4_K[16384, 4096]
-
blk.26.ffn_norm.biasF32[4096]
-
blk.26.ffn_norm.weightF32[4096]
-
blk.26.ffn_up.biasF32[16384]
-
blk.26.ffn_up.weightQ4_K[4096, 16384]
-
blk.27.attn_k_norm.biasF32[4096]
-
blk.27.attn_k_norm.weightF32[4096]
-
blk.27.attn_norm.biasF32[4096]
-
blk.27.attn_norm.weightF32[4096]
-
blk.27.attn_output.biasF32[4096]
-
blk.27.attn_output.weightQ4_K[4096, 4096]
-
blk.27.attn_q_norm.biasF32[4096]
-
blk.27.attn_q_norm.weightF32[4096]
-
blk.27.attn_qkv.biasF32[12288]
-
blk.27.attn_qkv.weightQ5_K[4096, 12288]
-
blk.27.ffn_down.biasF32[4096]
-
blk.27.ffn_down.weightQ6_K[16384, 4096]
-
blk.27.ffn_norm.biasF32[4096]
-
blk.27.ffn_norm.weightF32[4096]
-
blk.27.ffn_up.biasF32[16384]
-
blk.27.ffn_up.weightQ4_K[4096, 16384]
-
blk.28.attn_k_norm.biasF32[4096]
-
blk.28.attn_k_norm.weightF32[4096]
-
blk.28.attn_norm.biasF32[4096]
-
blk.28.attn_norm.weightF32[4096]
-
blk.28.attn_output.biasF32[4096]
-
blk.28.attn_output.weightQ4_K[4096, 4096]
-
blk.28.attn_q_norm.biasF32[4096]
-
blk.28.attn_q_norm.weightF32[4096]
-
blk.28.attn_qkv.biasF32[12288]
-
blk.28.attn_qkv.weightQ5_K[4096, 12288]
-
blk.28.ffn_down.biasF32[4096]
-
blk.28.ffn_down.weightQ6_K[16384, 4096]
-
blk.28.ffn_norm.biasF32[4096]
-
blk.28.ffn_norm.weightF32[4096]
-
blk.28.ffn_up.biasF32[16384]
-
blk.28.ffn_up.weightQ4_K[4096, 16384]
-
blk.29.attn_k_norm.biasF32[4096]
-
blk.29.attn_k_norm.weightF32[4096]
-
blk.29.attn_norm.biasF32[4096]
-
blk.29.attn_norm.weightF32[4096]
-
blk.29.attn_output.biasF32[4096]
-
blk.29.attn_output.weightQ4_K[4096, 4096]
-
blk.29.attn_q_norm.biasF32[4096]
-
blk.29.attn_q_norm.weightF32[4096]
-
blk.29.attn_qkv.biasF32[12288]
-
blk.29.attn_qkv.weightQ5_K[4096, 12288]
-
blk.29.ffn_down.biasF32[4096]
-
blk.29.ffn_down.weightQ6_K[16384, 4096]
-
blk.29.ffn_norm.biasF32[4096]
-
blk.29.ffn_norm.weightF32[4096]
-
blk.29.ffn_up.biasF32[16384]
-
blk.29.ffn_up.weightQ4_K[4096, 16384]
-
blk.30.attn_k_norm.biasF32[4096]
-
blk.30.attn_k_norm.weightF32[4096]
-
blk.30.attn_norm.biasF32[4096]
-
blk.30.attn_norm.weightF32[4096]
-
blk.30.attn_output.biasF32[4096]
-
blk.30.attn_output.weightQ4_K[4096, 4096]
-
blk.30.attn_q_norm.biasF32[4096]
-
blk.30.attn_q_norm.weightF32[4096]
-
blk.30.attn_qkv.biasF32[12288]
-
blk.30.attn_qkv.weightQ5_K[4096, 12288]
-
blk.30.ffn_down.biasF32[4096]
-
blk.30.ffn_down.weightQ6_K[16384, 4096]
-
blk.30.ffn_norm.biasF32[4096]
-
blk.30.ffn_norm.weightF32[4096]
-
blk.30.ffn_up.biasF32[16384]
-
blk.30.ffn_up.weightQ4_K[4096, 16384]
-
blk.31.attn_k_norm.biasF32[4096]
-
blk.31.attn_k_norm.weightF32[4096]
-
blk.31.attn_norm.biasF32[4096]
-
blk.31.attn_norm.weightF32[4096]
-
blk.31.attn_output.biasF32[4096]
-
blk.31.attn_output.weightQ4_K[4096, 4096]
-
blk.31.attn_q_norm.biasF32[4096]
-
blk.31.attn_q_norm.weightF32[4096]
-
blk.31.attn_qkv.biasF32[12288]
-
blk.31.attn_qkv.weightQ5_K[4096, 12288]
-
blk.31.ffn_down.biasF32[4096]
-
blk.31.ffn_down.weightQ6_K[16384, 4096]
-
blk.31.ffn_norm.biasF32[4096]
-
blk.31.ffn_norm.weightF32[4096]
-
blk.31.ffn_up.biasF32[16384]
-
blk.31.ffn_up.weightQ4_K[4096, 16384]
-
output_norm.biasF32[4096]
-
position_embd.weightF16[4096, 2048]
-
output_norm.weightF32[4096]
Metadata
Tensor
blk.0
blk.1
blk.2
blk.3
blk.4
blk.5
blk.6
blk.7
blk.8
blk.9
blk.10
blk.11
blk.12
blk.13
blk.14
blk.15
blk.16
blk.17
blk.18
blk.19
blk.20
blk.21
blk.22
blk.23
blk.24
blk.25
blk.26
blk.27
blk.28
blk.29
blk.30
blk.31