# Mistral-Small-3.2-24B-Instruct-pruned-Q3_K_M.gguf - GGUF Internal File Dump - Endian: LITTLE endian ## Key Value Metadata Store There are 49 key-value pairs in this file | POS | TYPE | Count | Key | Value | |----:|:---------|-------:|:---------------------------------------|:--------------------------------------------------------------------| | 1 | UINT32 | 1 | GGUF.version | 3 | | 2 | UINT64 | 1 | GGUF.tensor_count | 345 | | 3 | UINT64 | 1 | GGUF.kv_count | 46 | | 4 | STRING | 1 | general.architecture | `llama` | | 5 | STRING | 1 | general.type | `model` | | 6 | STRING | 1 | general.name | `Mistral Small 3.2 24B Instruct 2506` | | 7 | STRING | 1 | general.version | `2506` | | 8 | STRING | 1 | general.finetune | `Instruct` | | 9 | STRING | 1 | general.basename | `Mistral-Small-3.2` | | 10 | STRING | 1 | general.size_label | `24B` | | 11 | STRING | 1 | general.license | `apache-2.0` | | 12 | UINT32 | 1 | general.base_model.count | 1 | | 13 | STRING | 1 | general.base_model.0.name | `Mistral Small 3.1 24B Base 2503` | | 14 | STRING | 1 | general.base_model.0.version | `2503` | | 15 | STRING | 1 | general.base_model.0.organization | `Mistralai` | | 16 | STRING | 1 | general.base_model.0.repo_url | `https://huggingface.co/mistral`...`istral-Small-3.1-24B-Base-2503` | | 17 | [STRING] | 1 | general.tags | [ `image-text-to-text` ] | | 18 | [STRING] | 24 | general.languages | [ `en`, `fr`, `de`, `es`, `pt`, ... ] | | 19 | UINT32 | 1 | llama.context_length | 131072 | | 20 | UINT32 | 1 | llama.embedding_length | 5120 | | 21 | UINT32 | 1 | llama.feed_forward_length | 32768 | | 22 | UINT32 | 1 | llama.attention.head_count | 32 | | 23 | UINT32 | 1 | llama.attention.head_count_kv | 8 | | 24 | FLOAT32 | 1 | llama.rope.freq_base | 1000000000.0 | | 25 | FLOAT32 | 1 | llama.attention.layer_norm_rms_epsilon | 1e-05 | | 26 | UINT32 | 1 | llama.attention.key_length | 128 | | 27 | UINT32 | 1 | llama.attention.value_length | 128 | | 28 | UINT32 | 1 | llama.vocab_size | 131072 | | 29 | UINT32 | 1 | llama.rope.dimension_count | 128 | | 30 | STRING | 1 | tokenizer.ggml.model | `gpt2` | | 31 | STRING | 1 | tokenizer.ggml.pre | `tekken` | | 32 | [STRING] | 131072 | tokenizer.ggml.tokens | [ ``, ``, ``, `[INST]`, `[/INST]`, ... ] | | 33 | [INT32] | 131072 | tokenizer.ggml.token_type | [ 3, 3, 3, 3, 3, 3, 3, ... ] | | 34 | [STRING] | 269443 | tokenizer.ggml.merges | [ `Ġ Ġ`, `Ġ t`, `e r`, `i n`, `Ġ ĠĠĠ`, ... ] | | 35 | UINT32 | 1 | tokenizer.ggml.bos_token_id | 1 | | 36 | UINT32 | 1 | tokenizer.ggml.eos_token_id | 2 | | 37 | UINT32 | 1 | tokenizer.ggml.unknown_token_id | 0 | | 38 | UINT32 | 1 | tokenizer.ggml.padding_token_id | 11 | | 39 | BOOL | 1 | tokenizer.ggml.add_bos_token | True | | 40 | BOOL | 1 | tokenizer.ggml.add_sep_token | False | | 41 | BOOL | 1 | tokenizer.ggml.add_eos_token | False | | 42 | BOOL | 1 | tokenizer.ggml.add_space_prefix | False | | 43 | UINT32 | 1 | llama.block_count | 38 | | 44 | UINT32 | 1 | general.quantization_version | 2 | | 45 | UINT32 | 1 | general.file_type | 12 | | 46 | STRING | 1 | quantize.imatrix.file | `./imatrix/imatrix-Mistral-Smal`...`24B-Instruct-pruned-medium.dat` | | 47 | STRING | 1 | quantize.imatrix.dataset | `../../datasets/imatrix/text_eur_medium.txt` | | 48 | UINT32 | 1 | quantize.imatrix.entries_count | 266 | | 49 | UINT32 | 1 | quantize.imatrix.chunks_count | 1778 | ## Tensors Overview ~22B Elements Total number of elements in all tensors: 22460892160 Elements - [Mistral-Small-3.2-24B-Instruct-pruned-Q3\_K\_M.gguf - GGUF Internal File Dump](#mistral-small-32-24b-instruct-pruned-q3_k_mgguf---gguf-internal-file-dump) - [Key Value Metadata Store](#key-value-metadata-store) - [Tensors Overview ~22B Elements](#tensors-overview-22b-elements) - [Tensor Data Offset](#tensor-data-offset) - [Base Tensor Group : ~1B Elements](#base-tensor-group--1b-elements) - [Block 0 Tensor Group : ~556M Elements](#block-0-tensor-group--556m-elements) - [Block 1 Tensor Group : ~556M Elements](#block-1-tensor-group--556m-elements) - [Block 2 Tensor Group : ~556M Elements](#block-2-tensor-group--556m-elements) - [Block 3 Tensor Group : ~556M Elements](#block-3-tensor-group--556m-elements) - [Block 4 Tensor Group : ~556M Elements](#block-4-tensor-group--556m-elements) - [Block 5 Tensor Group : ~556M Elements](#block-5-tensor-group--556m-elements) - [Block 6 Tensor Group : ~556M Elements](#block-6-tensor-group--556m-elements) - [Block 7 Tensor Group : ~556M Elements](#block-7-tensor-group--556m-elements) - [Block 8 Tensor Group : ~556M Elements](#block-8-tensor-group--556m-elements) - [Block 9 Tensor Group : ~556M Elements](#block-9-tensor-group--556m-elements) - [Block 10 Tensor Group : ~556M Elements](#block-10-tensor-group--556m-elements) - [Block 11 Tensor Group : ~556M Elements](#block-11-tensor-group--556m-elements) - [Block 12 Tensor Group : ~556M Elements](#block-12-tensor-group--556m-elements) - [Block 13 Tensor Group : ~556M Elements](#block-13-tensor-group--556m-elements) - [Block 14 Tensor Group : ~556M Elements](#block-14-tensor-group--556m-elements) - [Block 15 Tensor Group : ~556M Elements](#block-15-tensor-group--556m-elements) - [Block 16 Tensor Group : ~556M Elements](#block-16-tensor-group--556m-elements) - [Block 17 Tensor Group : ~556M Elements](#block-17-tensor-group--556m-elements) - [Block 18 Tensor Group : ~556M Elements](#block-18-tensor-group--556m-elements) - [Block 19 Tensor Group : ~556M Elements](#block-19-tensor-group--556m-elements) - [Block 20 Tensor Group : ~556M Elements](#block-20-tensor-group--556m-elements) - [Block 21 Tensor Group : ~556M Elements](#block-21-tensor-group--556m-elements) - [Block 22 Tensor Group : ~556M Elements](#block-22-tensor-group--556m-elements) - [Block 23 Tensor Group : ~556M Elements](#block-23-tensor-group--556m-elements) - [Block 24 Tensor Group : ~556M Elements](#block-24-tensor-group--556m-elements) - [Block 25 Tensor Group : ~556M Elements](#block-25-tensor-group--556m-elements) - [Block 26 Tensor Group : ~556M Elements](#block-26-tensor-group--556m-elements) - [Block 27 Tensor Group : ~556M Elements](#block-27-tensor-group--556m-elements) - [Block 28 Tensor Group : ~556M Elements](#block-28-tensor-group--556m-elements) - [Block 29 Tensor Group : ~556M Elements](#block-29-tensor-group--556m-elements) - [Block 30 Tensor Group : ~556M Elements](#block-30-tensor-group--556m-elements) - [Block 31 Tensor Group : ~556M Elements](#block-31-tensor-group--556m-elements) - [Block 32 Tensor Group : ~556M Elements](#block-32-tensor-group--556m-elements) - [Block 33 Tensor Group : ~556M Elements](#block-33-tensor-group--556m-elements) - [Block 34 Tensor Group : ~556M Elements](#block-34-tensor-group--556m-elements) - [Block 35 Tensor Group : ~556M Elements](#block-35-tensor-group--556m-elements) - [Block 36 Tensor Group : ~556M Elements](#block-36-tensor-group--556m-elements) - [Block 37 Tensor Group : ~556M Elements](#block-37-tensor-group--556m-elements) ### Tensor Data Offset This table contains the offset and data segment relative to start of file | T_ID | Tensor Layer Name | Data Offset (B) | Data Size (B) | |-----:|:--------------------------|-----------------:|-----------------:| | 0 | output.weight | 0x7841e0 | 0x11300000 | | 1 | output_norm.weight | 0x11a841e0 | 0x5000 | | 2 | token_embd.weight | 0x11a891e0 | 0x11300000 | | 3 | blk.0.attn_k.weight | 0x22d891e0 | 0x226000 | | 4 | blk.0.attn_norm.weight | 0x22faf1e0 | 0x5000 | | 5 | blk.0.attn_output.weight | 0x22fb41e0 | 0xb40000 | | 6 | blk.0.attn_q.weight | 0x23af41e0 | 0x898000 | | 7 | blk.0.attn_v.weight | 0x2438c1e0 | 0x2d0000 | | 8 | blk.0.ffn_down.weight | 0x2465c1e0 | 0x5a00000 | | 9 | blk.0.ffn_gate.weight | 0x2a05c1e0 | 0x44c0000 | | 10 | blk.0.ffn_norm.weight | 0x2e51c1e0 | 0x5000 | | 11 | blk.0.ffn_up.weight | 0x2e5211e0 | 0x44c0000 | | 12 | blk.1.attn_k.weight | 0x329e11e0 | 0x226000 | | 13 | blk.1.attn_norm.weight | 0x32c071e0 | 0x5000 | | 14 | blk.1.attn_output.weight | 0x32c0c1e0 | 0xb40000 | | 15 | blk.1.attn_q.weight | 0x3374c1e0 | 0x898000 | | 16 | blk.1.attn_v.weight | 0x33fe41e0 | 0x2d0000 | | 17 | blk.1.ffn_down.weight | 0x342b41e0 | 0x5a00000 | | 18 | blk.1.ffn_gate.weight | 0x39cb41e0 | 0x44c0000 | | 19 | blk.1.ffn_norm.weight | 0x3e1741e0 | 0x5000 | | 20 | blk.1.ffn_up.weight | 0x3e1791e0 | 0x44c0000 | | 21 | blk.2.attn_k.weight | 0x426391e0 | 0x226000 | | 22 | blk.2.attn_norm.weight | 0x4285f1e0 | 0x5000 | | 23 | blk.2.attn_output.weight | 0x428641e0 | 0xb40000 | | 24 | blk.2.attn_q.weight | 0x433a41e0 | 0x898000 | | 25 | blk.2.attn_v.weight | 0x43c3c1e0 | 0x2d0000 | | 26 | blk.2.ffn_down.weight | 0x43f0c1e0 | 0x5a00000 | | 27 | blk.2.ffn_gate.weight | 0x4990c1e0 | 0x44c0000 | | 28 | blk.2.ffn_norm.weight | 0x4ddcc1e0 | 0x5000 | | 29 | blk.2.ffn_up.weight | 0x4ddd11e0 | 0x44c0000 | | 30 | blk.3.attn_k.weight | 0x522911e0 | 0x226000 | | 31 | blk.3.attn_norm.weight | 0x524b71e0 | 0x5000 | | 32 | blk.3.attn_output.weight | 0x524bc1e0 | 0xb40000 | | 33 | blk.3.attn_q.weight | 0x52ffc1e0 | 0x898000 | | 34 | blk.3.attn_v.weight | 0x538941e0 | 0x2d0000 | | 35 | blk.3.ffn_down.weight | 0x53b641e0 | 0x5a00000 | | 36 | blk.3.ffn_gate.weight | 0x595641e0 | 0x44c0000 | | 37 | blk.3.ffn_norm.weight | 0x5da241e0 | 0x5000 | | 38 | blk.3.ffn_up.weight | 0x5da291e0 | 0x44c0000 | | 39 | blk.4.attn_k.weight | 0x61ee91e0 | 0x226000 | | 40 | blk.4.attn_norm.weight | 0x6210f1e0 | 0x5000 | | 41 | blk.4.attn_output.weight | 0x621141e0 | 0xb40000 | | 42 | blk.4.attn_q.weight | 0x62c541e0 | 0x898000 | | 43 | blk.4.attn_v.weight | 0x634ec1e0 | 0x2d0000 | | 44 | blk.4.ffn_down.weight | 0x637bc1e0 | 0x5a00000 | | 45 | blk.4.ffn_gate.weight | 0x691bc1e0 | 0x44c0000 | | 46 | blk.4.ffn_norm.weight | 0x6d67c1e0 | 0x5000 | | 47 | blk.4.ffn_up.weight | 0x6d6811e0 | 0x44c0000 | | 48 | blk.5.attn_k.weight | 0x71b411e0 | 0x226000 | | 49 | blk.5.attn_norm.weight | 0x71d671e0 | 0x5000 | | 50 | blk.5.attn_output.weight | 0x71d6c1e0 | 0xb40000 | | 51 | blk.5.attn_q.weight | 0x728ac1e0 | 0x898000 | | 52 | blk.5.attn_v.weight | 0x731441e0 | 0x2d0000 | | 53 | blk.5.ffn_down.weight | 0x734141e0 | 0x5a00000 | | 54 | blk.5.ffn_gate.weight | 0x78e141e0 | 0x44c0000 | | 55 | blk.5.ffn_norm.weight | 0x7d2d41e0 | 0x5000 | | 56 | blk.5.ffn_up.weight | 0x7d2d91e0 | 0x44c0000 | | 57 | blk.6.attn_k.weight | 0x817991e0 | 0x226000 | | 58 | blk.6.attn_norm.weight | 0x819bf1e0 | 0x5000 | | 59 | blk.6.attn_output.weight | 0x819c41e0 | 0xb40000 | | 60 | blk.6.attn_q.weight | 0x825041e0 | 0x898000 | | 61 | blk.6.attn_v.weight | 0x82d9c1e0 | 0x2d0000 | | 62 | blk.6.ffn_down.weight | 0x8306c1e0 | 0x5a00000 | | 63 | blk.6.ffn_gate.weight | 0x88a6c1e0 | 0x44c0000 | | 64 | blk.6.ffn_norm.weight | 0x8cf2c1e0 | 0x5000 | | 65 | blk.6.ffn_up.weight | 0x8cf311e0 | 0x44c0000 | | 66 | blk.7.attn_k.weight | 0x913f11e0 | 0x226000 | | 67 | blk.7.attn_norm.weight | 0x916171e0 | 0x5000 | | 68 | blk.7.attn_output.weight | 0x9161c1e0 | 0xb40000 | | 69 | blk.7.attn_q.weight | 0x9215c1e0 | 0x898000 | | 70 | blk.7.attn_v.weight | 0x929f41e0 | 0x2d0000 | | 71 | blk.7.ffn_down.weight | 0x92cc41e0 | 0x5a00000 | | 72 | blk.7.ffn_gate.weight | 0x986c41e0 | 0x44c0000 | | 73 | blk.7.ffn_norm.weight | 0x9cb841e0 | 0x5000 | | 74 | blk.7.ffn_up.weight | 0x9cb891e0 | 0x44c0000 | | 75 | blk.8.attn_k.weight | 0xa10491e0 | 0x226000 | | 76 | blk.8.attn_norm.weight | 0xa126f1e0 | 0x5000 | | 77 | blk.8.attn_output.weight | 0xa12741e0 | 0xb40000 | | 78 | blk.8.attn_q.weight | 0xa1db41e0 | 0x898000 | | 79 | blk.8.attn_v.weight | 0xa264c1e0 | 0x2d0000 | | 80 | blk.8.ffn_down.weight | 0xa291c1e0 | 0x5a00000 | | 81 | blk.8.ffn_gate.weight | 0xa831c1e0 | 0x44c0000 | | 82 | blk.8.ffn_norm.weight | 0xac7dc1e0 | 0x5000 | | 83 | blk.8.ffn_up.weight | 0xac7e11e0 | 0x44c0000 | | 84 | blk.9.attn_k.weight | 0xb0ca11e0 | 0x226000 | | 85 | blk.9.attn_norm.weight | 0xb0ec71e0 | 0x5000 | | 86 | blk.9.attn_output.weight | 0xb0ecc1e0 | 0xb40000 | | 87 | blk.9.attn_q.weight | 0xb1a0c1e0 | 0x898000 | | 88 | blk.9.attn_v.weight | 0xb22a41e0 | 0x2d0000 | | 89 | blk.9.ffn_down.weight | 0xb25741e0 | 0x5a00000 | | 90 | blk.9.ffn_gate.weight | 0xb7f741e0 | 0x44c0000 | | 91 | blk.9.ffn_norm.weight | 0xbc4341e0 | 0x5000 | | 92 | blk.9.ffn_up.weight | 0xbc4391e0 | 0x44c0000 | | 93 | blk.10.attn_k.weight | 0xc08f91e0 | 0x226000 | | 94 | blk.10.attn_norm.weight | 0xc0b1f1e0 | 0x5000 | | 95 | blk.10.attn_output.weight | 0xc0b241e0 | 0xb40000 | | 96 | blk.10.attn_q.weight | 0xc16641e0 | 0x898000 | | 97 | blk.10.attn_v.weight | 0xc1efc1e0 | 0x2d0000 | | 98 | blk.10.ffn_down.weight | 0xc21cc1e0 | 0x5a00000 | | 99 | blk.10.ffn_gate.weight | 0xc7bcc1e0 | 0x44c0000 | | 100 | blk.10.ffn_norm.weight | 0xcc08c1e0 | 0x5000 | | 101 | blk.10.ffn_up.weight | 0xcc0911e0 | 0x44c0000 | | 102 | blk.11.attn_k.weight | 0xd05511e0 | 0x226000 | | 103 | blk.11.attn_norm.weight | 0xd07771e0 | 0x5000 | | 104 | blk.11.attn_output.weight | 0xd077c1e0 | 0xb40000 | | 105 | blk.11.attn_q.weight | 0xd12bc1e0 | 0x898000 | | 106 | blk.11.attn_v.weight | 0xd1b541e0 | 0x2d0000 | | 107 | blk.11.ffn_down.weight | 0xd1e241e0 | 0x5a00000 | | 108 | blk.11.ffn_gate.weight | 0xd78241e0 | 0x44c0000 | | 109 | blk.11.ffn_norm.weight | 0xdbce41e0 | 0x5000 | | 110 | blk.11.ffn_up.weight | 0xdbce91e0 | 0x44c0000 | | 111 | blk.12.attn_k.weight | 0xe01a91e0 | 0x226000 | | 112 | blk.12.attn_norm.weight | 0xe03cf1e0 | 0x5000 | | 113 | blk.12.attn_output.weight | 0xe03d41e0 | 0xb40000 | | 114 | blk.12.attn_q.weight | 0xe0f141e0 | 0x898000 | | 115 | blk.12.attn_v.weight | 0xe17ac1e0 | 0x2d0000 | | 116 | blk.12.ffn_down.weight | 0xe1a7c1e0 | 0x5a00000 | | 117 | blk.12.ffn_gate.weight | 0xe747c1e0 | 0x44c0000 | | 118 | blk.12.ffn_norm.weight | 0xeb93c1e0 | 0x5000 | | 119 | blk.12.ffn_up.weight | 0xeb9411e0 | 0x44c0000 | | 120 | blk.13.attn_k.weight | 0xefe011e0 | 0x226000 | | 121 | blk.13.attn_norm.weight | 0xf00271e0 | 0x5000 | | 122 | blk.13.attn_output.weight | 0xf002c1e0 | 0xb40000 | | 123 | blk.13.attn_q.weight | 0xf0b6c1e0 | 0x898000 | | 124 | blk.13.attn_v.weight | 0xf14041e0 | 0x2d0000 | | 125 | blk.13.ffn_down.weight | 0xf16d41e0 | 0x5a00000 | | 126 | blk.13.ffn_gate.weight | 0xf70d41e0 | 0x44c0000 | | 127 | blk.13.ffn_norm.weight | 0xfb5941e0 | 0x5000 | | 128 | blk.13.ffn_up.weight | 0xfb5991e0 | 0x44c0000 | | 129 | blk.14.attn_k.weight | 0xffa591e0 | 0x226000 | | 130 | blk.14.attn_norm.weight | 0xffc7f1e0 | 0x5000 | | 131 | blk.14.attn_output.weight | 0xffc841e0 | 0xb40000 | | 132 | blk.14.attn_q.weight | 0x1007c41e0 | 0x898000 | | 133 | blk.14.attn_v.weight | 0x10105c1e0 | 0x2d0000 | | 134 | blk.14.ffn_down.weight | 0x10132c1e0 | 0x5a00000 | | 135 | blk.14.ffn_gate.weight | 0x106d2c1e0 | 0x44c0000 | | 136 | blk.14.ffn_norm.weight | 0x10b1ec1e0 | 0x5000 | | 137 | blk.14.ffn_up.weight | 0x10b1f11e0 | 0x44c0000 | | 138 | blk.15.attn_k.weight | 0x10f6b11e0 | 0x226000 | | 139 | blk.15.attn_norm.weight | 0x10f8d71e0 | 0x5000 | | 140 | blk.15.attn_output.weight | 0x10f8dc1e0 | 0xb40000 | | 141 | blk.15.attn_q.weight | 0x11041c1e0 | 0x898000 | | 142 | blk.15.attn_v.weight | 0x110cb41e0 | 0x2d0000 | | 143 | blk.15.ffn_down.weight | 0x110f841e0 | 0x5a00000 | | 144 | blk.15.ffn_gate.weight | 0x1169841e0 | 0x44c0000 | | 145 | blk.15.ffn_norm.weight | 0x11ae441e0 | 0x5000 | | 146 | blk.15.ffn_up.weight | 0x11ae491e0 | 0x44c0000 | | 147 | blk.16.attn_k.weight | 0x11f3091e0 | 0x226000 | | 148 | blk.16.attn_norm.weight | 0x11f52f1e0 | 0x5000 | | 149 | blk.16.attn_output.weight | 0x11f5341e0 | 0xb40000 | | 150 | blk.16.attn_q.weight | 0x1200741e0 | 0x898000 | | 151 | blk.16.attn_v.weight | 0x12090c1e0 | 0x2d0000 | | 152 | blk.16.ffn_down.weight | 0x120bdc1e0 | 0x5a00000 | | 153 | blk.16.ffn_gate.weight | 0x1265dc1e0 | 0x44c0000 | | 154 | blk.16.ffn_norm.weight | 0x12aa9c1e0 | 0x5000 | | 155 | blk.16.ffn_up.weight | 0x12aaa11e0 | 0x44c0000 | | 156 | blk.17.attn_k.weight | 0x12ef611e0 | 0x1a4000 | | 157 | blk.17.attn_norm.weight | 0x12f1051e0 | 0x5000 | | 158 | blk.17.attn_output.weight | 0x12f10a1e0 | 0xb40000 | | 159 | blk.17.attn_q.weight | 0x12fc4a1e0 | 0x690000 | | 160 | blk.17.attn_v.weight | 0x1302da1e0 | 0x226000 | | 161 | blk.17.ffn_down.weight | 0x1305001e0 | 0x5a00000 | | 162 | blk.17.ffn_gate.weight | 0x135f001e0 | 0x44c0000 | | 163 | blk.17.ffn_norm.weight | 0x13a3c01e0 | 0x5000 | | 164 | blk.17.ffn_up.weight | 0x13a3c51e0 | 0x44c0000 | | 165 | blk.18.attn_k.weight | 0x13e8851e0 | 0x1a4000 | | 166 | blk.18.attn_norm.weight | 0x13ea291e0 | 0x5000 | | 167 | blk.18.attn_output.weight | 0x13ea2e1e0 | 0xb40000 | | 168 | blk.18.attn_q.weight | 0x13f56e1e0 | 0x690000 | | 169 | blk.18.attn_v.weight | 0x13fbfe1e0 | 0x226000 | | 170 | blk.18.ffn_down.weight | 0x13fe241e0 | 0x5a00000 | | 171 | blk.18.ffn_gate.weight | 0x1458241e0 | 0x44c0000 | | 172 | blk.18.ffn_norm.weight | 0x149ce41e0 | 0x5000 | | 173 | blk.18.ffn_up.weight | 0x149ce91e0 | 0x44c0000 | | 174 | blk.19.attn_k.weight | 0x14e1a91e0 | 0x226000 | | 175 | blk.19.attn_norm.weight | 0x14e3cf1e0 | 0x5000 | | 176 | blk.19.attn_output.weight | 0x14e3d41e0 | 0xb40000 | | 177 | blk.19.attn_q.weight | 0x14ef141e0 | 0x898000 | | 178 | blk.19.attn_v.weight | 0x14f7ac1e0 | 0x2d0000 | | 179 | blk.19.ffn_down.weight | 0x14fa7c1e0 | 0x5a00000 | | 180 | blk.19.ffn_gate.weight | 0x15547c1e0 | 0x44c0000 | | 181 | blk.19.ffn_norm.weight | 0x15993c1e0 | 0x5000 | | 182 | blk.19.ffn_up.weight | 0x1599411e0 | 0x44c0000 | | 183 | blk.20.attn_k.weight | 0x15de011e0 | 0x1a4000 | | 184 | blk.20.attn_norm.weight | 0x15dfa51e0 | 0x5000 | | 185 | blk.20.attn_output.weight | 0x15dfaa1e0 | 0xb40000 | | 186 | blk.20.attn_q.weight | 0x15eaea1e0 | 0x690000 | | 187 | blk.20.attn_v.weight | 0x15f17a1e0 | 0x226000 | | 188 | blk.20.ffn_down.weight | 0x15f3a01e0 | 0x5a00000 | | 189 | blk.20.ffn_gate.weight | 0x164da01e0 | 0x3480000 | | 190 | blk.20.ffn_norm.weight | 0x1682201e0 | 0x5000 | | 191 | blk.20.ffn_up.weight | 0x1682251e0 | 0x3480000 | | 192 | blk.21.attn_k.weight | 0x16b6a51e0 | 0x226000 | | 193 | blk.21.attn_norm.weight | 0x16b8cb1e0 | 0x5000 | | 194 | blk.21.attn_output.weight | 0x16b8d01e0 | 0xb40000 | | 195 | blk.21.attn_q.weight | 0x16c4101e0 | 0x898000 | | 196 | blk.21.attn_v.weight | 0x16cca81e0 | 0x2d0000 | | 197 | blk.21.ffn_down.weight | 0x16cf781e0 | 0x5a00000 | | 198 | blk.21.ffn_gate.weight | 0x1729781e0 | 0x3480000 | | 199 | blk.21.ffn_norm.weight | 0x175df81e0 | 0x5000 | | 200 | blk.21.ffn_up.weight | 0x175dfd1e0 | 0x3480000 | | 201 | blk.22.attn_k.weight | 0x17927d1e0 | 0x1a4000 | | 202 | blk.22.attn_norm.weight | 0x1794211e0 | 0x5000 | | 203 | blk.22.attn_output.weight | 0x1794261e0 | 0xb40000 | | 204 | blk.22.attn_q.weight | 0x179f661e0 | 0x690000 | | 205 | blk.22.attn_v.weight | 0x17a5f61e0 | 0x226000 | | 206 | blk.22.ffn_down.weight | 0x17a81c1e0 | 0x5a00000 | | 207 | blk.22.ffn_gate.weight | 0x18021c1e0 | 0x3480000 | | 208 | blk.22.ffn_norm.weight | 0x18369c1e0 | 0x5000 | | 209 | blk.22.ffn_up.weight | 0x1836a11e0 | 0x3480000 | | 210 | blk.23.attn_k.weight | 0x186b211e0 | 0x1a4000 | | 211 | blk.23.attn_norm.weight | 0x186cc51e0 | 0x5000 | | 212 | blk.23.attn_output.weight | 0x186cca1e0 | 0xb40000 | | 213 | blk.23.attn_q.weight | 0x18780a1e0 | 0x690000 | | 214 | blk.23.attn_v.weight | 0x187e9a1e0 | 0x226000 | | 215 | blk.23.ffn_down.weight | 0x1880c01e0 | 0x5a00000 | | 216 | blk.23.ffn_gate.weight | 0x18dac01e0 | 0x3480000 | | 217 | blk.23.ffn_norm.weight | 0x190f401e0 | 0x5000 | | 218 | blk.23.ffn_up.weight | 0x190f451e0 | 0x3480000 | | 219 | blk.24.attn_k.weight | 0x1943c51e0 | 0x1a4000 | | 220 | blk.24.attn_norm.weight | 0x1945691e0 | 0x5000 | | 221 | blk.24.attn_output.weight | 0x19456e1e0 | 0xb40000 | | 222 | blk.24.attn_q.weight | 0x1950ae1e0 | 0x690000 | | 223 | blk.24.attn_v.weight | 0x19573e1e0 | 0x226000 | | 224 | blk.24.ffn_down.weight | 0x1959641e0 | 0x5a00000 | | 225 | blk.24.ffn_gate.weight | 0x19b3641e0 | 0x3480000 | | 226 | blk.24.ffn_norm.weight | 0x19e7e41e0 | 0x5000 | | 227 | blk.24.ffn_up.weight | 0x19e7e91e0 | 0x3480000 | | 228 | blk.25.attn_k.weight | 0x1a1c691e0 | 0x1a4000 | | 229 | blk.25.attn_norm.weight | 0x1a1e0d1e0 | 0x5000 | | 230 | blk.25.attn_output.weight | 0x1a1e121e0 | 0xb40000 | | 231 | blk.25.attn_q.weight | 0x1a29521e0 | 0x690000 | | 232 | blk.25.attn_v.weight | 0x1a2fe21e0 | 0x226000 | | 233 | blk.25.ffn_down.weight | 0x1a32081e0 | 0x5a00000 | | 234 | blk.25.ffn_gate.weight | 0x1a8c081e0 | 0x3480000 | | 235 | blk.25.ffn_norm.weight | 0x1ac0881e0 | 0x5000 | | 236 | blk.25.ffn_up.weight | 0x1ac08d1e0 | 0x3480000 | | 237 | blk.26.attn_k.weight | 0x1af50d1e0 | 0x1a4000 | | 238 | blk.26.attn_norm.weight | 0x1af6b11e0 | 0x5000 | | 239 | blk.26.attn_output.weight | 0x1af6b61e0 | 0xb40000 | | 240 | blk.26.attn_q.weight | 0x1b01f61e0 | 0x690000 | | 241 | blk.26.attn_v.weight | 0x1b08861e0 | 0x226000 | | 242 | blk.26.ffn_down.weight | 0x1b0aac1e0 | 0x5a00000 | | 243 | blk.26.ffn_gate.weight | 0x1b64ac1e0 | 0x3480000 | | 244 | blk.26.ffn_norm.weight | 0x1b992c1e0 | 0x5000 | | 245 | blk.26.ffn_up.weight | 0x1b99311e0 | 0x3480000 | | 246 | blk.27.attn_k.weight | 0x1bcdb11e0 | 0x226000 | | 247 | blk.27.attn_norm.weight | 0x1bcfd71e0 | 0x5000 | | 248 | blk.27.attn_output.weight | 0x1bcfdc1e0 | 0xb40000 | | 249 | blk.27.attn_q.weight | 0x1bdb1c1e0 | 0x898000 | | 250 | blk.27.attn_v.weight | 0x1be3b41e0 | 0x2d0000 | | 251 | blk.27.ffn_down.weight | 0x1be6841e0 | 0x5a00000 | | 252 | blk.27.ffn_gate.weight | 0x1c40841e0 | 0x3480000 | | 253 | blk.27.ffn_norm.weight | 0x1c75041e0 | 0x5000 | | 254 | blk.27.ffn_up.weight | 0x1c75091e0 | 0x3480000 | | 255 | blk.28.attn_k.weight | 0x1ca9891e0 | 0x1a4000 | | 256 | blk.28.attn_norm.weight | 0x1cab2d1e0 | 0x5000 | | 257 | blk.28.attn_output.weight | 0x1cab321e0 | 0xb40000 | | 258 | blk.28.attn_q.weight | 0x1cb6721e0 | 0x690000 | | 259 | blk.28.attn_v.weight | 0x1cbd021e0 | 0x226000 | | 260 | blk.28.ffn_down.weight | 0x1cbf281e0 | 0x5a00000 | | 261 | blk.28.ffn_gate.weight | 0x1d19281e0 | 0x3480000 | | 262 | blk.28.ffn_norm.weight | 0x1d4da81e0 | 0x5000 | | 263 | blk.28.ffn_up.weight | 0x1d4dad1e0 | 0x3480000 | | 264 | blk.29.attn_k.weight | 0x1d822d1e0 | 0x1a4000 | | 265 | blk.29.attn_norm.weight | 0x1d83d11e0 | 0x5000 | | 266 | blk.29.attn_output.weight | 0x1d83d61e0 | 0xb40000 | | 267 | blk.29.attn_q.weight | 0x1d8f161e0 | 0x690000 | | 268 | blk.29.attn_v.weight | 0x1d95a61e0 | 0x226000 | | 269 | blk.29.ffn_down.weight | 0x1d97cc1e0 | 0x5a00000 | | 270 | blk.29.ffn_gate.weight | 0x1df1cc1e0 | 0x3480000 | | 271 | blk.29.ffn_norm.weight | 0x1e264c1e0 | 0x5000 | | 272 | blk.29.ffn_up.weight | 0x1e26511e0 | 0x3480000 | | 273 | blk.30.attn_k.weight | 0x1e5ad11e0 | 0x1a4000 | | 274 | blk.30.attn_norm.weight | 0x1e5c751e0 | 0x5000 | | 275 | blk.30.attn_output.weight | 0x1e5c7a1e0 | 0xb40000 | | 276 | blk.30.attn_q.weight | 0x1e67ba1e0 | 0x690000 | | 277 | blk.30.attn_v.weight | 0x1e6e4a1e0 | 0x226000 | | 278 | blk.30.ffn_down.weight | 0x1e70701e0 | 0x5a00000 | | 279 | blk.30.ffn_gate.weight | 0x1eca701e0 | 0x3480000 | | 280 | blk.30.ffn_norm.weight | 0x1efef01e0 | 0x5000 | | 281 | blk.30.ffn_up.weight | 0x1efef51e0 | 0x3480000 | | 282 | blk.31.attn_k.weight | 0x1f33751e0 | 0x1a4000 | | 283 | blk.31.attn_norm.weight | 0x1f35191e0 | 0x5000 | | 284 | blk.31.attn_output.weight | 0x1f351e1e0 | 0xb40000 | | 285 | blk.31.attn_q.weight | 0x1f405e1e0 | 0x690000 | | 286 | blk.31.attn_v.weight | 0x1f46ee1e0 | 0x226000 | | 287 | blk.31.ffn_down.weight | 0x1f49141e0 | 0x5a00000 | | 288 | blk.31.ffn_gate.weight | 0x1fa3141e0 | 0x3480000 | | 289 | blk.31.ffn_norm.weight | 0x1fd7941e0 | 0x5000 | | 290 | blk.31.ffn_up.weight | 0x1fd7991e0 | 0x3480000 | | 291 | blk.32.attn_k.weight | 0x200c191e0 | 0x1a4000 | | 292 | blk.32.attn_norm.weight | 0x200dbd1e0 | 0x5000 | | 293 | blk.32.attn_output.weight | 0x200dc21e0 | 0xb40000 | | 294 | blk.32.attn_q.weight | 0x2019021e0 | 0x690000 | | 295 | blk.32.attn_v.weight | 0x201f921e0 | 0x226000 | | 296 | blk.32.ffn_down.weight | 0x2021b81e0 | 0x5a00000 | | 297 | blk.32.ffn_gate.weight | 0x207bb81e0 | 0x3480000 | | 298 | blk.32.ffn_norm.weight | 0x20b0381e0 | 0x5000 | | 299 | blk.32.ffn_up.weight | 0x20b03d1e0 | 0x3480000 | | 300 | blk.33.attn_k.weight | 0x20e4bd1e0 | 0x1a4000 | | 301 | blk.33.attn_norm.weight | 0x20e6611e0 | 0x5000 | | 302 | blk.33.attn_output.weight | 0x20e6661e0 | 0xb40000 | | 303 | blk.33.attn_q.weight | 0x20f1a61e0 | 0x690000 | | 304 | blk.33.attn_v.weight | 0x20f8361e0 | 0x226000 | | 305 | blk.33.ffn_down.weight | 0x20fa5c1e0 | 0x5a00000 | | 306 | blk.33.ffn_gate.weight | 0x21545c1e0 | 0x3480000 | | 307 | blk.33.ffn_norm.weight | 0x2188dc1e0 | 0x5000 | | 308 | blk.33.ffn_up.weight | 0x2188e11e0 | 0x3480000 | | 309 | blk.34.attn_k.weight | 0x21bd611e0 | 0x1a4000 | | 310 | blk.34.attn_norm.weight | 0x21bf051e0 | 0x5000 | | 311 | blk.34.attn_output.weight | 0x21bf0a1e0 | 0xb40000 | | 312 | blk.34.attn_q.weight | 0x21ca4a1e0 | 0x690000 | | 313 | blk.34.attn_v.weight | 0x21d0da1e0 | 0x226000 | | 314 | blk.34.ffn_down.weight | 0x21d3001e0 | 0x5a00000 | | 315 | blk.34.ffn_gate.weight | 0x222d001e0 | 0x3480000 | | 316 | blk.34.ffn_norm.weight | 0x2261801e0 | 0x5000 | | 317 | blk.34.ffn_up.weight | 0x2261851e0 | 0x3480000 | | 318 | blk.35.attn_k.weight | 0x2296051e0 | 0x1a4000 | | 319 | blk.35.attn_norm.weight | 0x2297a91e0 | 0x5000 | | 320 | blk.35.attn_output.weight | 0x2297ae1e0 | 0xb40000 | | 321 | blk.35.attn_q.weight | 0x22a2ee1e0 | 0x690000 | | 322 | blk.35.attn_v.weight | 0x22a97e1e0 | 0x226000 | | 323 | blk.35.ffn_down.weight | 0x22aba41e0 | 0x5a00000 | | 324 | blk.35.ffn_gate.weight | 0x2305a41e0 | 0x3480000 | | 325 | blk.35.ffn_norm.weight | 0x233a241e0 | 0x5000 | | 326 | blk.35.ffn_up.weight | 0x233a291e0 | 0x3480000 | | 327 | blk.36.attn_k.weight | 0x236ea91e0 | 0x1a4000 | | 328 | blk.36.attn_norm.weight | 0x23704d1e0 | 0x5000 | | 329 | blk.36.attn_output.weight | 0x2370521e0 | 0xb40000 | | 330 | blk.36.attn_q.weight | 0x237b921e0 | 0x690000 | | 331 | blk.36.attn_v.weight | 0x2382221e0 | 0x226000 | | 332 | blk.36.ffn_down.weight | 0x2384481e0 | 0x5a00000 | | 333 | blk.36.ffn_gate.weight | 0x23de481e0 | 0x3480000 | | 334 | blk.36.ffn_norm.weight | 0x2412c81e0 | 0x5000 | | 335 | blk.36.ffn_up.weight | 0x2412cd1e0 | 0x3480000 | | 336 | blk.37.attn_k.weight | 0x24474d1e0 | 0x1a4000 | | 337 | blk.37.attn_norm.weight | 0x2448f11e0 | 0x5000 | | 338 | blk.37.attn_output.weight | 0x2448f61e0 | 0xb40000 | | 339 | blk.37.attn_q.weight | 0x2454361e0 | 0x690000 | | 340 | blk.37.attn_v.weight | 0x245ac61e0 | 0x226000 | | 341 | blk.37.ffn_down.weight | 0x245cec1e0 | 0x5a00000 | | 342 | blk.37.ffn_gate.weight | 0x24b6ec1e0 | 0x3480000 | | 343 | blk.37.ffn_norm.weight | 0x24eb6c1e0 | 0x5000 | | 344 | blk.37.ffn_up.weight | 0x24eb711e0 | 0x3480000 | ### Base Tensor Group : ~1B Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:-------------------|:---------------------------------|:------------------|:----------------------|:-----|--------:| | 0 | output.weight | Output (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | Q3_K | 3.4375 | | 1 | output_norm.weight | Output Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 2 | token_embd.weight | Token Embedding (W) | (~671M) 671088640 | 5120 x 131072 x 1 x 1 | Q3_K | 3.4375 | - Total elements in base: ( ~1B) 1342182400 - Percentage of total elements: 5.98% - Bits per Weight (BPW) for base: 3.4376 bits ### Block 0 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----|--------:| | 3 | blk.0.attn_k.weight | Block 0 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 4 | blk.0.attn_norm.weight | Block 0 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 5 | blk.0.attn_output.weight | Block 0 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 6 | blk.0.attn_q.weight | Block 0 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | 3.4375 | | 7 | blk.0.attn_v.weight | Block 0 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | 4.5000 | | 8 | blk.0.ffn_down.weight | Block 0 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 9 | blk.0.ffn_gate.weight | Block 0 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | | 10 | blk.0.ffn_norm.weight | Block 0 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 11 | blk.0.ffn_up.weight | Block 0 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | - Total elements in blk.0: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.0: 3.8089 bits ### Block 1 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----|--------:| | 12 | blk.1.attn_k.weight | Block 1 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 13 | blk.1.attn_norm.weight | Block 1 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 14 | blk.1.attn_output.weight | Block 1 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 15 | blk.1.attn_q.weight | Block 1 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | 3.4375 | | 16 | blk.1.attn_v.weight | Block 1 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | 4.5000 | | 17 | blk.1.ffn_down.weight | Block 1 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 18 | blk.1.ffn_gate.weight | Block 1 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | | 19 | blk.1.ffn_norm.weight | Block 1 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 20 | blk.1.ffn_up.weight | Block 1 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | - Total elements in blk.1: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.1: 3.8089 bits ### Block 2 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----|--------:| | 21 | blk.2.attn_k.weight | Block 2 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 22 | blk.2.attn_norm.weight | Block 2 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 23 | blk.2.attn_output.weight | Block 2 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 24 | blk.2.attn_q.weight | Block 2 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | 3.4375 | | 25 | blk.2.attn_v.weight | Block 2 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | 4.5000 | | 26 | blk.2.ffn_down.weight | Block 2 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 27 | blk.2.ffn_gate.weight | Block 2 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | | 28 | blk.2.ffn_norm.weight | Block 2 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 29 | blk.2.ffn_up.weight | Block 2 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | - Total elements in blk.2: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.2: 3.8089 bits ### Block 3 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----|--------:| | 30 | blk.3.attn_k.weight | Block 3 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 31 | blk.3.attn_norm.weight | Block 3 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 32 | blk.3.attn_output.weight | Block 3 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 33 | blk.3.attn_q.weight | Block 3 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | 3.4375 | | 34 | blk.3.attn_v.weight | Block 3 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | 4.5000 | | 35 | blk.3.ffn_down.weight | Block 3 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 36 | blk.3.ffn_gate.weight | Block 3 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | | 37 | blk.3.ffn_norm.weight | Block 3 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 38 | blk.3.ffn_up.weight | Block 3 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | - Total elements in blk.3: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.3: 3.8089 bits ### Block 4 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----|--------:| | 39 | blk.4.attn_k.weight | Block 4 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 40 | blk.4.attn_norm.weight | Block 4 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 41 | blk.4.attn_output.weight | Block 4 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 42 | blk.4.attn_q.weight | Block 4 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | 3.4375 | | 43 | blk.4.attn_v.weight | Block 4 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | 4.5000 | | 44 | blk.4.ffn_down.weight | Block 4 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 45 | blk.4.ffn_gate.weight | Block 4 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | | 46 | blk.4.ffn_norm.weight | Block 4 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 47 | blk.4.ffn_up.weight | Block 4 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | - Total elements in blk.4: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.4: 3.8089 bits ### Block 5 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----|--------:| | 48 | blk.5.attn_k.weight | Block 5 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 49 | blk.5.attn_norm.weight | Block 5 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 50 | blk.5.attn_output.weight | Block 5 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 51 | blk.5.attn_q.weight | Block 5 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | 3.4375 | | 52 | blk.5.attn_v.weight | Block 5 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | 4.5000 | | 53 | blk.5.ffn_down.weight | Block 5 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 54 | blk.5.ffn_gate.weight | Block 5 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | | 55 | blk.5.ffn_norm.weight | Block 5 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 56 | blk.5.ffn_up.weight | Block 5 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | - Total elements in blk.5: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.5: 3.8089 bits ### Block 6 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----|--------:| | 57 | blk.6.attn_k.weight | Block 6 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 58 | blk.6.attn_norm.weight | Block 6 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 59 | blk.6.attn_output.weight | Block 6 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 60 | blk.6.attn_q.weight | Block 6 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | 3.4375 | | 61 | blk.6.attn_v.weight | Block 6 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | 4.5000 | | 62 | blk.6.ffn_down.weight | Block 6 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 63 | blk.6.ffn_gate.weight | Block 6 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | | 64 | blk.6.ffn_norm.weight | Block 6 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 65 | blk.6.ffn_up.weight | Block 6 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | - Total elements in blk.6: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.6: 3.8089 bits ### Block 7 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----|--------:| | 66 | blk.7.attn_k.weight | Block 7 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 67 | blk.7.attn_norm.weight | Block 7 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 68 | blk.7.attn_output.weight | Block 7 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 69 | blk.7.attn_q.weight | Block 7 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | 3.4375 | | 70 | blk.7.attn_v.weight | Block 7 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | 4.5000 | | 71 | blk.7.ffn_down.weight | Block 7 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 72 | blk.7.ffn_gate.weight | Block 7 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | | 73 | blk.7.ffn_norm.weight | Block 7 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 74 | blk.7.ffn_up.weight | Block 7 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | - Total elements in blk.7: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.7: 3.8089 bits ### Block 8 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----|--------:| | 75 | blk.8.attn_k.weight | Block 8 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 76 | blk.8.attn_norm.weight | Block 8 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 77 | blk.8.attn_output.weight | Block 8 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 78 | blk.8.attn_q.weight | Block 8 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | 3.4375 | | 79 | blk.8.attn_v.weight | Block 8 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | 4.5000 | | 80 | blk.8.ffn_down.weight | Block 8 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 81 | blk.8.ffn_gate.weight | Block 8 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | | 82 | blk.8.ffn_norm.weight | Block 8 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 83 | blk.8.ffn_up.weight | Block 8 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | - Total elements in blk.8: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.8: 3.8089 bits ### Block 9 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:-------------------------|:-----------------------------------------------|:------------------|:----------------------|:-----|--------:| | 84 | blk.9.attn_k.weight | Block 9 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 85 | blk.9.attn_norm.weight | Block 9 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 86 | blk.9.attn_output.weight | Block 9 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 87 | blk.9.attn_q.weight | Block 9 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | 3.4375 | | 88 | blk.9.attn_v.weight | Block 9 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | 4.5000 | | 89 | blk.9.ffn_down.weight | Block 9 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 90 | blk.9.ffn_gate.weight | Block 9 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | | 91 | blk.9.ffn_norm.weight | Block 9 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 92 | blk.9.ffn_up.weight | Block 9 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | - Total elements in blk.9: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.9: 3.8089 bits ### Block 10 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 93 | blk.10.attn_k.weight | Block 10 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 94 | blk.10.attn_norm.weight | Block 10 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 95 | blk.10.attn_output.weight | Block 10 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 96 | blk.10.attn_q.weight | Block 10 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | 3.4375 | | 97 | blk.10.attn_v.weight | Block 10 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | 4.5000 | | 98 | blk.10.ffn_down.weight | Block 10 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 99 | blk.10.ffn_gate.weight | Block 10 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | | 100 | blk.10.ffn_norm.weight | Block 10 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 101 | blk.10.ffn_up.weight | Block 10 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | - Total elements in blk.10: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.10: 3.8089 bits ### Block 11 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 102 | blk.11.attn_k.weight | Block 11 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 103 | blk.11.attn_norm.weight | Block 11 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 104 | blk.11.attn_output.weight | Block 11 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 105 | blk.11.attn_q.weight | Block 11 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | 3.4375 | | 106 | blk.11.attn_v.weight | Block 11 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | 4.5000 | | 107 | blk.11.ffn_down.weight | Block 11 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 108 | blk.11.ffn_gate.weight | Block 11 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | | 109 | blk.11.ffn_norm.weight | Block 11 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 110 | blk.11.ffn_up.weight | Block 11 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | - Total elements in blk.11: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.11: 3.8089 bits ### Block 12 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 111 | blk.12.attn_k.weight | Block 12 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 112 | blk.12.attn_norm.weight | Block 12 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 113 | blk.12.attn_output.weight | Block 12 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 114 | blk.12.attn_q.weight | Block 12 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | 3.4375 | | 115 | blk.12.attn_v.weight | Block 12 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | 4.5000 | | 116 | blk.12.ffn_down.weight | Block 12 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 117 | blk.12.ffn_gate.weight | Block 12 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | | 118 | blk.12.ffn_norm.weight | Block 12 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 119 | blk.12.ffn_up.weight | Block 12 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | - Total elements in blk.12: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.12: 3.8089 bits ### Block 13 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 120 | blk.13.attn_k.weight | Block 13 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 121 | blk.13.attn_norm.weight | Block 13 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 122 | blk.13.attn_output.weight | Block 13 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 123 | blk.13.attn_q.weight | Block 13 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | 3.4375 | | 124 | blk.13.attn_v.weight | Block 13 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | 4.5000 | | 125 | blk.13.ffn_down.weight | Block 13 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 126 | blk.13.ffn_gate.weight | Block 13 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | | 127 | blk.13.ffn_norm.weight | Block 13 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 128 | blk.13.ffn_up.weight | Block 13 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | - Total elements in blk.13: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.13: 3.8089 bits ### Block 14 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 129 | blk.14.attn_k.weight | Block 14 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 130 | blk.14.attn_norm.weight | Block 14 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 131 | blk.14.attn_output.weight | Block 14 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 132 | blk.14.attn_q.weight | Block 14 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | 3.4375 | | 133 | blk.14.attn_v.weight | Block 14 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | 4.5000 | | 134 | blk.14.ffn_down.weight | Block 14 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 135 | blk.14.ffn_gate.weight | Block 14 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | | 136 | blk.14.ffn_norm.weight | Block 14 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 137 | blk.14.ffn_up.weight | Block 14 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | - Total elements in blk.14: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.14: 3.8089 bits ### Block 15 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 138 | blk.15.attn_k.weight | Block 15 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 139 | blk.15.attn_norm.weight | Block 15 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 140 | blk.15.attn_output.weight | Block 15 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 141 | blk.15.attn_q.weight | Block 15 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | 3.4375 | | 142 | blk.15.attn_v.weight | Block 15 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | 4.5000 | | 143 | blk.15.ffn_down.weight | Block 15 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 144 | blk.15.ffn_gate.weight | Block 15 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | | 145 | blk.15.ffn_norm.weight | Block 15 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 146 | blk.15.ffn_up.weight | Block 15 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | - Total elements in blk.15: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.15: 3.8089 bits ### Block 16 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 147 | blk.16.attn_k.weight | Block 16 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 148 | blk.16.attn_norm.weight | Block 16 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 149 | blk.16.attn_output.weight | Block 16 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 150 | blk.16.attn_q.weight | Block 16 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | 3.4375 | | 151 | blk.16.attn_v.weight | Block 16 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | 4.5000 | | 152 | blk.16.ffn_down.weight | Block 16 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 153 | blk.16.ffn_gate.weight | Block 16 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | | 154 | blk.16.ffn_norm.weight | Block 16 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 155 | blk.16.ffn_up.weight | Block 16 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | - Total elements in blk.16: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.16: 3.8089 bits ### Block 17 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 156 | blk.17.attn_k.weight | Block 17 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | 2.6250 | | 157 | blk.17.attn_norm.weight | Block 17 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 158 | blk.17.attn_output.weight | Block 17 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 159 | blk.17.attn_q.weight | Block 17 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | 2.6250 | | 160 | blk.17.attn_v.weight | Block 17 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 161 | blk.17.ffn_down.weight | Block 17 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 162 | blk.17.ffn_gate.weight | Block 17 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | | 163 | blk.17.ffn_norm.weight | Block 17 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 164 | blk.17.ffn_up.weight | Block 17 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | - Total elements in blk.17: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.17: 3.7605 bits ### Block 18 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 165 | blk.18.attn_k.weight | Block 18 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | 2.6250 | | 166 | blk.18.attn_norm.weight | Block 18 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 167 | blk.18.attn_output.weight | Block 18 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 168 | blk.18.attn_q.weight | Block 18 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | 2.6250 | | 169 | blk.18.attn_v.weight | Block 18 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 170 | blk.18.ffn_down.weight | Block 18 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 171 | blk.18.ffn_gate.weight | Block 18 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | | 172 | blk.18.ffn_norm.weight | Block 18 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 173 | blk.18.ffn_up.weight | Block 18 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | - Total elements in blk.18: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.18: 3.7605 bits ### Block 19 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 174 | blk.19.attn_k.weight | Block 19 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 175 | blk.19.attn_norm.weight | Block 19 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 176 | blk.19.attn_output.weight | Block 19 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 177 | blk.19.attn_q.weight | Block 19 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | 3.4375 | | 178 | blk.19.attn_v.weight | Block 19 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | 4.5000 | | 179 | blk.19.ffn_down.weight | Block 19 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 180 | blk.19.ffn_gate.weight | Block 19 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | | 181 | blk.19.ffn_norm.weight | Block 19 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 182 | blk.19.ffn_up.weight | Block 19 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q3_K | 3.4375 | - Total elements in blk.19: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.19: 3.8089 bits ### Block 20 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 183 | blk.20.attn_k.weight | Block 20 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | 2.6250 | | 184 | blk.20.attn_norm.weight | Block 20 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 185 | blk.20.attn_output.weight | Block 20 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 186 | blk.20.attn_q.weight | Block 20 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | 2.6250 | | 187 | blk.20.attn_v.weight | Block 20 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 188 | blk.20.ffn_down.weight | Block 20 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 189 | blk.20.ffn_gate.weight | Block 20 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | | 190 | blk.20.ffn_norm.weight | Block 20 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 191 | blk.20.ffn_up.weight | Block 20 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | - Total elements in blk.20: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.20: 3.2700 bits ### Block 21 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 192 | blk.21.attn_k.weight | Block 21 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 193 | blk.21.attn_norm.weight | Block 21 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 194 | blk.21.attn_output.weight | Block 21 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 195 | blk.21.attn_q.weight | Block 21 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | 3.4375 | | 196 | blk.21.attn_v.weight | Block 21 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | 4.5000 | | 197 | blk.21.ffn_down.weight | Block 21 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 198 | blk.21.ffn_gate.weight | Block 21 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | | 199 | blk.21.ffn_norm.weight | Block 21 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 200 | blk.21.ffn_up.weight | Block 21 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | - Total elements in blk.21: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.21: 3.3183 bits ### Block 22 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 201 | blk.22.attn_k.weight | Block 22 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | 2.6250 | | 202 | blk.22.attn_norm.weight | Block 22 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 203 | blk.22.attn_output.weight | Block 22 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 204 | blk.22.attn_q.weight | Block 22 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | 2.6250 | | 205 | blk.22.attn_v.weight | Block 22 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 206 | blk.22.ffn_down.weight | Block 22 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 207 | blk.22.ffn_gate.weight | Block 22 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | | 208 | blk.22.ffn_norm.weight | Block 22 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 209 | blk.22.ffn_up.weight | Block 22 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | - Total elements in blk.22: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.22: 3.2700 bits ### Block 23 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 210 | blk.23.attn_k.weight | Block 23 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | 2.6250 | | 211 | blk.23.attn_norm.weight | Block 23 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 212 | blk.23.attn_output.weight | Block 23 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 213 | blk.23.attn_q.weight | Block 23 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | 2.6250 | | 214 | blk.23.attn_v.weight | Block 23 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 215 | blk.23.ffn_down.weight | Block 23 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 216 | blk.23.ffn_gate.weight | Block 23 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | | 217 | blk.23.ffn_norm.weight | Block 23 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 218 | blk.23.ffn_up.weight | Block 23 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | - Total elements in blk.23: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.23: 3.2700 bits ### Block 24 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 219 | blk.24.attn_k.weight | Block 24 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | 2.6250 | | 220 | blk.24.attn_norm.weight | Block 24 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 221 | blk.24.attn_output.weight | Block 24 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 222 | blk.24.attn_q.weight | Block 24 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | 2.6250 | | 223 | blk.24.attn_v.weight | Block 24 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 224 | blk.24.ffn_down.weight | Block 24 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 225 | blk.24.ffn_gate.weight | Block 24 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | | 226 | blk.24.ffn_norm.weight | Block 24 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 227 | blk.24.ffn_up.weight | Block 24 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | - Total elements in blk.24: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.24: 3.2700 bits ### Block 25 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 228 | blk.25.attn_k.weight | Block 25 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | 2.6250 | | 229 | blk.25.attn_norm.weight | Block 25 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 230 | blk.25.attn_output.weight | Block 25 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 231 | blk.25.attn_q.weight | Block 25 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | 2.6250 | | 232 | blk.25.attn_v.weight | Block 25 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 233 | blk.25.ffn_down.weight | Block 25 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 234 | blk.25.ffn_gate.weight | Block 25 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | | 235 | blk.25.ffn_norm.weight | Block 25 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 236 | blk.25.ffn_up.weight | Block 25 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | - Total elements in blk.25: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.25: 3.2700 bits ### Block 26 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 237 | blk.26.attn_k.weight | Block 26 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | 2.6250 | | 238 | blk.26.attn_norm.weight | Block 26 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 239 | blk.26.attn_output.weight | Block 26 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 240 | blk.26.attn_q.weight | Block 26 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | 2.6250 | | 241 | blk.26.attn_v.weight | Block 26 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 242 | blk.26.ffn_down.weight | Block 26 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 243 | blk.26.ffn_gate.weight | Block 26 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | | 244 | blk.26.ffn_norm.weight | Block 26 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 245 | blk.26.ffn_up.weight | Block 26 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | - Total elements in blk.26: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.26: 3.2700 bits ### Block 27 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 246 | blk.27.attn_k.weight | Block 27 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 247 | blk.27.attn_norm.weight | Block 27 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 248 | blk.27.attn_output.weight | Block 27 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 249 | blk.27.attn_q.weight | Block 27 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q3_K | 3.4375 | | 250 | blk.27.attn_v.weight | Block 27 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q4_K | 4.5000 | | 251 | blk.27.ffn_down.weight | Block 27 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 252 | blk.27.ffn_gate.weight | Block 27 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | | 253 | blk.27.ffn_norm.weight | Block 27 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 254 | blk.27.ffn_up.weight | Block 27 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | - Total elements in blk.27: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.27: 3.3183 bits ### Block 28 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 255 | blk.28.attn_k.weight | Block 28 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | 2.6250 | | 256 | blk.28.attn_norm.weight | Block 28 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 257 | blk.28.attn_output.weight | Block 28 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 258 | blk.28.attn_q.weight | Block 28 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | 2.6250 | | 259 | blk.28.attn_v.weight | Block 28 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 260 | blk.28.ffn_down.weight | Block 28 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 261 | blk.28.ffn_gate.weight | Block 28 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | | 262 | blk.28.ffn_norm.weight | Block 28 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 263 | blk.28.ffn_up.weight | Block 28 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | - Total elements in blk.28: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.28: 3.2700 bits ### Block 29 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 264 | blk.29.attn_k.weight | Block 29 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | 2.6250 | | 265 | blk.29.attn_norm.weight | Block 29 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 266 | blk.29.attn_output.weight | Block 29 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 267 | blk.29.attn_q.weight | Block 29 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | 2.6250 | | 268 | blk.29.attn_v.weight | Block 29 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 269 | blk.29.ffn_down.weight | Block 29 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 270 | blk.29.ffn_gate.weight | Block 29 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | | 271 | blk.29.ffn_norm.weight | Block 29 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 272 | blk.29.ffn_up.weight | Block 29 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | - Total elements in blk.29: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.29: 3.2700 bits ### Block 30 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 273 | blk.30.attn_k.weight | Block 30 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | 2.6250 | | 274 | blk.30.attn_norm.weight | Block 30 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 275 | blk.30.attn_output.weight | Block 30 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 276 | blk.30.attn_q.weight | Block 30 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | 2.6250 | | 277 | blk.30.attn_v.weight | Block 30 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 278 | blk.30.ffn_down.weight | Block 30 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 279 | blk.30.ffn_gate.weight | Block 30 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | | 280 | blk.30.ffn_norm.weight | Block 30 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 281 | blk.30.ffn_up.weight | Block 30 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | - Total elements in blk.30: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.30: 3.2700 bits ### Block 31 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 282 | blk.31.attn_k.weight | Block 31 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | 2.6250 | | 283 | blk.31.attn_norm.weight | Block 31 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 284 | blk.31.attn_output.weight | Block 31 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 285 | blk.31.attn_q.weight | Block 31 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | 2.6250 | | 286 | blk.31.attn_v.weight | Block 31 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 287 | blk.31.ffn_down.weight | Block 31 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 288 | blk.31.ffn_gate.weight | Block 31 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | | 289 | blk.31.ffn_norm.weight | Block 31 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 290 | blk.31.ffn_up.weight | Block 31 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | - Total elements in blk.31: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.31: 3.2700 bits ### Block 32 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 291 | blk.32.attn_k.weight | Block 32 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | 2.6250 | | 292 | blk.32.attn_norm.weight | Block 32 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 293 | blk.32.attn_output.weight | Block 32 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 294 | blk.32.attn_q.weight | Block 32 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | 2.6250 | | 295 | blk.32.attn_v.weight | Block 32 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 296 | blk.32.ffn_down.weight | Block 32 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 297 | blk.32.ffn_gate.weight | Block 32 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | | 298 | blk.32.ffn_norm.weight | Block 32 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 299 | blk.32.ffn_up.weight | Block 32 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | - Total elements in blk.32: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.32: 3.2700 bits ### Block 33 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 300 | blk.33.attn_k.weight | Block 33 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | 2.6250 | | 301 | blk.33.attn_norm.weight | Block 33 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 302 | blk.33.attn_output.weight | Block 33 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 303 | blk.33.attn_q.weight | Block 33 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | 2.6250 | | 304 | blk.33.attn_v.weight | Block 33 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 305 | blk.33.ffn_down.weight | Block 33 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 306 | blk.33.ffn_gate.weight | Block 33 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | | 307 | blk.33.ffn_norm.weight | Block 33 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 308 | blk.33.ffn_up.weight | Block 33 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | - Total elements in blk.33: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.33: 3.2700 bits ### Block 34 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 309 | blk.34.attn_k.weight | Block 34 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | 2.6250 | | 310 | blk.34.attn_norm.weight | Block 34 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 311 | blk.34.attn_output.weight | Block 34 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 312 | blk.34.attn_q.weight | Block 34 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | 2.6250 | | 313 | blk.34.attn_v.weight | Block 34 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 314 | blk.34.ffn_down.weight | Block 34 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 315 | blk.34.ffn_gate.weight | Block 34 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | | 316 | blk.34.ffn_norm.weight | Block 34 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 317 | blk.34.ffn_up.weight | Block 34 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | - Total elements in blk.34: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.34: 3.2700 bits ### Block 35 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 318 | blk.35.attn_k.weight | Block 35 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | 2.6250 | | 319 | blk.35.attn_norm.weight | Block 35 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 320 | blk.35.attn_output.weight | Block 35 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 321 | blk.35.attn_q.weight | Block 35 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | 2.6250 | | 322 | blk.35.attn_v.weight | Block 35 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 323 | blk.35.ffn_down.weight | Block 35 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 324 | blk.35.ffn_gate.weight | Block 35 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | | 325 | blk.35.ffn_norm.weight | Block 35 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 326 | blk.35.ffn_up.weight | Block 35 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | - Total elements in blk.35: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.35: 3.2700 bits ### Block 36 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 327 | blk.36.attn_k.weight | Block 36 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | 2.6250 | | 328 | blk.36.attn_norm.weight | Block 36 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 329 | blk.36.attn_output.weight | Block 36 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 330 | blk.36.attn_q.weight | Block 36 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | 2.6250 | | 331 | blk.36.attn_v.weight | Block 36 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 332 | blk.36.ffn_down.weight | Block 36 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 333 | blk.36.ffn_gate.weight | Block 36 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | | 334 | blk.36.ffn_norm.weight | Block 36 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 335 | blk.36.ffn_up.weight | Block 36 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | - Total elements in blk.36: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.36: 3.2700 bits ### Block 37 Tensor Group : ~556M Elements | T_ID | Tensor Layer Name | Human Friendly Tensor Layer Name | Elements | Shape | Type | BPW | |-----:|:--------------------------|:------------------------------------------------|:------------------|:----------------------|:-----|--------:| | 336 | blk.37.attn_k.weight | Block 37 Attention Key (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q2_K | 2.6250 | | 337 | blk.37.attn_norm.weight | Block 37 Attention Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 338 | blk.37.attn_output.weight | Block 37 Attention Output (W) | ( ~21M) 20971520 | 4096 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 339 | blk.37.attn_q.weight | Block 37 Attention Query (W) | ( ~21M) 20971520 | 5120 x 4096 x 1 x 1 | Q2_K | 2.6250 | | 340 | blk.37.attn_v.weight | Block 37 Attention Value (W) | ( ~5M) 5242880 | 5120 x 1024 x 1 x 1 | Q3_K | 3.4375 | | 341 | blk.37.ffn_down.weight | Block 37 Feed-Forward Network "Down" (W) | (~168M) 167772160 | 32768 x 5120 x 1 x 1 | Q4_K | 4.5000 | | 342 | blk.37.ffn_gate.weight | Block 37 Feed-Forward Network "Gate" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | | 343 | blk.37.ffn_norm.weight | Block 37 Feed-Forward Network Normalization (W) | ( ~5K) 5120 | 5120 x 1 x 1 x 1 | F32 | 32.0000 | | 344 | blk.37.ffn_up.weight | Block 37 Feed-Forward Network "Up" (W) | (~168M) 167772160 | 5120 x 32768 x 1 x 1 | Q2_K | 2.6250 | - Total elements in blk.37: (~556M) 555755520 - Percentage of total elements: 2.47% - Bits per Weight (BPW) for blk.37: 3.2700 bits Total BPW for Mistral-Small-3.2-24B-Instruct-pruned-Q3_K_M.gguf: 3.5467 bits