legraphista
commited on
Commit
•
bb72316
1
Parent(s):
e773c40
Upload imatrix.log with huggingface_hub
Browse files- imatrix.log +144 -0
imatrix.log
ADDED
@@ -0,0 +1,144 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
llama_model_loader: loaded meta data with 24 key-value pairs and 291 tensors from internlm2_5-7b-chat-IMat-GGUF/internlm2_5-7b-chat.Q8_0.gguf.hardlink.gguf (version GGUF V3 (latest))
|
2 |
+
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
|
3 |
+
llama_model_loader: - kv 0: general.architecture str = internlm2
|
4 |
+
llama_model_loader: - kv 1: general.name str = InternLM2
|
5 |
+
llama_model_loader: - kv 2: internlm2.context_length u32 = 32768
|
6 |
+
llama_model_loader: - kv 3: internlm2.block_count u32 = 32
|
7 |
+
llama_model_loader: - kv 4: internlm2.embedding_length u32 = 4096
|
8 |
+
llama_model_loader: - kv 5: internlm2.feed_forward_length u32 = 14336
|
9 |
+
llama_model_loader: - kv 6: internlm2.rope.freq_base f32 = 1000000.000000
|
10 |
+
llama_model_loader: - kv 7: internlm2.attention.head_count u32 = 32
|
11 |
+
llama_model_loader: - kv 8: internlm2.attention.layer_norm_rms_epsilon f32 = 0.000010
|
12 |
+
llama_model_loader: - kv 9: internlm2.attention.head_count_kv u32 = 8
|
13 |
+
llama_model_loader: - kv 10: general.file_type u32 = 7
|
14 |
+
llama_model_loader: - kv 11: tokenizer.ggml.model str = llama
|
15 |
+
llama_model_loader: - kv 12: tokenizer.ggml.pre str = default
|
16 |
+
llama_model_loader: - kv 13: tokenizer.ggml.tokens arr[str,92544] = ["<unk>", "<s>", "</s>", "<0x00>", "<...
|
17 |
+
llama_model_loader: - kv 14: tokenizer.ggml.scores arr[f32,92544] = [0.000000, 0.000000, 0.000000, 0.0000...
|
18 |
+
llama_model_loader: - kv 15: tokenizer.ggml.token_type arr[i32,92544] = [2, 3, 3, 6, 6, 6, 6, 6, 6, 6, 6, 6, ...
|
19 |
+
llama_model_loader: - kv 16: tokenizer.ggml.add_space_prefix bool = false
|
20 |
+
llama_model_loader: - kv 17: tokenizer.ggml.bos_token_id u32 = 1
|
21 |
+
llama_model_loader: - kv 18: tokenizer.ggml.eos_token_id u32 = 92542
|
22 |
+
llama_model_loader: - kv 19: tokenizer.ggml.padding_token_id u32 = 2
|
23 |
+
llama_model_loader: - kv 20: tokenizer.ggml.add_bos_token bool = true
|
24 |
+
llama_model_loader: - kv 21: tokenizer.ggml.add_eos_token bool = false
|
25 |
+
llama_model_loader: - kv 22: tokenizer.chat_template str = {{ bos_token }}{% for message in mess...
|
26 |
+
llama_model_loader: - kv 23: general.quantization_version u32 = 2
|
27 |
+
llama_model_loader: - type f32: 65 tensors
|
28 |
+
llama_model_loader: - type q8_0: 226 tensors
|
29 |
+
llm_load_vocab: special tokens cache size = 259
|
30 |
+
llm_load_vocab: token to piece cache size = 0.5531 MB
|
31 |
+
llm_load_print_meta: format = GGUF V3 (latest)
|
32 |
+
llm_load_print_meta: arch = internlm2
|
33 |
+
llm_load_print_meta: vocab type = SPM
|
34 |
+
llm_load_print_meta: n_vocab = 92544
|
35 |
+
llm_load_print_meta: n_merges = 0
|
36 |
+
llm_load_print_meta: n_ctx_train = 32768
|
37 |
+
llm_load_print_meta: n_embd = 4096
|
38 |
+
llm_load_print_meta: n_head = 32
|
39 |
+
llm_load_print_meta: n_head_kv = 8
|
40 |
+
llm_load_print_meta: n_layer = 32
|
41 |
+
llm_load_print_meta: n_rot = 128
|
42 |
+
llm_load_print_meta: n_swa = 0
|
43 |
+
llm_load_print_meta: n_embd_head_k = 128
|
44 |
+
llm_load_print_meta: n_embd_head_v = 128
|
45 |
+
llm_load_print_meta: n_gqa = 4
|
46 |
+
llm_load_print_meta: n_embd_k_gqa = 1024
|
47 |
+
llm_load_print_meta: n_embd_v_gqa = 1024
|
48 |
+
llm_load_print_meta: f_norm_eps = 0.0e+00
|
49 |
+
llm_load_print_meta: f_norm_rms_eps = 1.0e-05
|
50 |
+
llm_load_print_meta: f_clamp_kqv = 0.0e+00
|
51 |
+
llm_load_print_meta: f_max_alibi_bias = 0.0e+00
|
52 |
+
llm_load_print_meta: f_logit_scale = 0.0e+00
|
53 |
+
llm_load_print_meta: n_ff = 14336
|
54 |
+
llm_load_print_meta: n_expert = 0
|
55 |
+
llm_load_print_meta: n_expert_used = 0
|
56 |
+
llm_load_print_meta: causal attn = 1
|
57 |
+
llm_load_print_meta: pooling type = 0
|
58 |
+
llm_load_print_meta: rope type = 0
|
59 |
+
llm_load_print_meta: rope scaling = linear
|
60 |
+
llm_load_print_meta: freq_base_train = 1000000.0
|
61 |
+
llm_load_print_meta: freq_scale_train = 1
|
62 |
+
llm_load_print_meta: n_ctx_orig_yarn = 32768
|
63 |
+
llm_load_print_meta: rope_finetuned = unknown
|
64 |
+
llm_load_print_meta: ssm_d_conv = 0
|
65 |
+
llm_load_print_meta: ssm_d_inner = 0
|
66 |
+
llm_load_print_meta: ssm_d_state = 0
|
67 |
+
llm_load_print_meta: ssm_dt_rank = 0
|
68 |
+
llm_load_print_meta: model type = 7B
|
69 |
+
llm_load_print_meta: model ftype = Q8_0
|
70 |
+
llm_load_print_meta: model params = 7.74 B
|
71 |
+
llm_load_print_meta: model size = 7.66 GiB (8.50 BPW)
|
72 |
+
llm_load_print_meta: general.name = InternLM2
|
73 |
+
llm_load_print_meta: BOS token = 1 '<s>'
|
74 |
+
llm_load_print_meta: EOS token = 92542 '[UNUSED_TOKEN_145]'
|
75 |
+
llm_load_print_meta: UNK token = 0 '<unk>'
|
76 |
+
llm_load_print_meta: PAD token = 2 '</s>'
|
77 |
+
llm_load_print_meta: LF token = 13 '<0x0A>'
|
78 |
+
llm_load_print_meta: max token length = 384
|
79 |
+
ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no
|
80 |
+
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
|
81 |
+
ggml_cuda_init: found 1 CUDA devices:
|
82 |
+
Device 0: NVIDIA GeForce RTX 4090, compute capability 8.9, VMM: yes
|
83 |
+
llm_load_tensors: ggml ctx size = 0.27 MiB
|
84 |
+
llm_load_tensors: offloading 32 repeating layers to GPU
|
85 |
+
llm_load_tensors: offloading non-repeating layers to GPU
|
86 |
+
llm_load_tensors: offloaded 33/33 layers to GPU
|
87 |
+
llm_load_tensors: CPU buffer size = 384.09 MiB
|
88 |
+
llm_load_tensors: CUDA0 buffer size = 7457.11 MiB
|
89 |
+
.............................................................................................
|
90 |
+
llama_new_context_with_model: n_ctx = 512
|
91 |
+
llama_new_context_with_model: n_batch = 512
|
92 |
+
llama_new_context_with_model: n_ubatch = 512
|
93 |
+
llama_new_context_with_model: flash_attn = 0
|
94 |
+
llama_new_context_with_model: freq_base = 1000000.0
|
95 |
+
llama_new_context_with_model: freq_scale = 1
|
96 |
+
llama_kv_cache_init: CUDA0 KV buffer size = 64.00 MiB
|
97 |
+
llama_new_context_with_model: KV self size = 64.00 MiB, K (f16): 32.00 MiB, V (f16): 32.00 MiB
|
98 |
+
llama_new_context_with_model: CUDA_Host output buffer size = 0.35 MiB
|
99 |
+
llama_new_context_with_model: CUDA0 compute buffer size = 188.75 MiB
|
100 |
+
llama_new_context_with_model: CUDA_Host compute buffer size = 9.01 MiB
|
101 |
+
llama_new_context_with_model: graph nodes = 1030
|
102 |
+
llama_new_context_with_model: graph splits = 2
|
103 |
+
|
104 |
+
system_info: n_threads = 25 / 32 | AVX = 1 | AVX_VNNI = 0 | AVX2 = 1 | AVX512 = 1 | AVX512_VBMI = 1 | AVX512_VNNI = 1 | AVX512_BF16 = 1 | FMA = 1 | NEON = 0 | SVE = 0 | ARM_FMA = 0 | F16C = 1 | FP16_VA = 0 | WASM_SIMD = 0 | BLAS = 1 | SSE3 = 1 | SSSE3 = 1 | VSX = 0 | MATMUL_INT8 = 0 | LLAMAFILE = 1 |
|
105 |
+
compute_imatrix: tokenizing the input ..
|
106 |
+
compute_imatrix: tokenization took 125.116 ms
|
107 |
+
compute_imatrix: computing over 136 chunks with batch_size 512
|
108 |
+
compute_imatrix: 1.50 seconds per pass - ETA 3.38 minutes
|
109 |
+
[1]5.3660,[2]3.9648,[3]3.7793,[4]4.3816,[5]4.2898,[6]3.9196,[7]4.5676,[8]4.6225,[9]5.0903,
|
110 |
+
save_imatrix: stored collected data after 10 chunks in internlm2_5-7b-chat-IMat-GGUF/imatrix.dat
|
111 |
+
[10]5.2353,[11]4.8014,[12]5.1504,[13]5.7994,[14]6.1277,[15]6.6241,[16]6.8574,[17]6.5020,[18]6.7127,[19]7.0476,
|
112 |
+
save_imatrix: stored collected data after 20 chunks in internlm2_5-7b-chat-IMat-GGUF/imatrix.dat
|
113 |
+
[20]6.8084,[21]6.8750,[22]7.0183,[23]7.0552,[24]7.0645,[25]7.2564,[26]7.4545,[27]7.6265,[28]7.6720,[29]7.8119,
|
114 |
+
save_imatrix: stored collected data after 30 chunks in internlm2_5-7b-chat-IMat-GGUF/imatrix.dat
|
115 |
+
[30]8.0432,[31]8.0995,[32]7.8028,[33]7.4882,[34]7.1976,[35]6.9321,[36]6.7823,[37]6.6661,[38]6.5867,[39]6.5069,
|
116 |
+
save_imatrix: stored collected data after 40 chunks in internlm2_5-7b-chat-IMat-GGUF/imatrix.dat
|
117 |
+
[40]6.4153,[41]6.3593,[42]6.2493,[43]6.2304,[44]6.2903,[45]6.3623,[46]6.4987,[47]6.4758,[48]6.7113,[49]6.8952,
|
118 |
+
save_imatrix: stored collected data after 50 chunks in internlm2_5-7b-chat-IMat-GGUF/imatrix.dat
|
119 |
+
[50]7.0713,[51]7.1993,[52]7.3858,[53]7.2623,[54]7.3476,[55]7.4306,[56]7.5495,[57]7.4182,[58]7.4262,[59]7.4570,
|
120 |
+
save_imatrix: stored collected data after 60 chunks in internlm2_5-7b-chat-IMat-GGUF/imatrix.dat
|
121 |
+
[60]7.5600,[61]7.6967,[62]7.8142,[63]7.8778,[64]7.8814,[65]7.8901,[66]7.8617,[67]7.8262,[68]7.7442,[69]7.7165,
|
122 |
+
save_imatrix: stored collected data after 70 chunks in internlm2_5-7b-chat-IMat-GGUF/imatrix.dat
|
123 |
+
[70]7.7870,[71]7.8071,[72]7.7318,[73]7.6962,[74]7.6986,[75]7.6555,[76]7.6464,[77]7.6257,[78]7.6434,[79]7.5874,
|
124 |
+
save_imatrix: stored collected data after 80 chunks in internlm2_5-7b-chat-IMat-GGUF/imatrix.dat
|
125 |
+
[80]7.5887,[81]7.5350,[82]7.4994,[83]7.4532,[84]7.4326,[85]7.3718,[86]7.3389,[87]7.3023,[88]7.3406,[89]7.3531,
|
126 |
+
save_imatrix: stored collected data after 90 chunks in internlm2_5-7b-chat-IMat-GGUF/imatrix.dat
|
127 |
+
[90]7.3153,[91]7.3366,[92]7.3520,[93]7.3018,[94]7.2910,[95]7.2756,[96]7.3042,[97]7.3015,[98]7.3043,[99]7.2553,
|
128 |
+
save_imatrix: stored collected data after 100 chunks in internlm2_5-7b-chat-IMat-GGUF/imatrix.dat
|
129 |
+
[100]7.2220,[101]7.1726,[102]7.1216,[103]7.0830,[104]7.0406,[105]7.0004,[106]6.9762,[107]6.9796,[108]7.0178,[109]7.0906,
|
130 |
+
save_imatrix: stored collected data after 110 chunks in internlm2_5-7b-chat-IMat-GGUF/imatrix.dat
|
131 |
+
[110]7.1629,[111]7.2254,[112]7.3286,[113]7.3967,[114]7.4209,[115]7.4120,[116]7.4278,[117]7.4226,[118]7.4174,[119]7.3764,
|
132 |
+
save_imatrix: stored collected data after 120 chunks in internlm2_5-7b-chat-IMat-GGUF/imatrix.dat
|
133 |
+
[120]7.3639,[121]7.3932,[122]7.3968,[123]7.3999,[124]7.4179,[125]7.4566,[126]7.4840,[127]7.4967,[128]7.5167,[129]7.5438,
|
134 |
+
save_imatrix: stored collected data after 130 chunks in internlm2_5-7b-chat-IMat-GGUF/imatrix.dat
|
135 |
+
[130]7.4995,[131]7.5409,[132]7.6049,[133]7.6445,[134]7.7057,[135]7.7517,[136]7.8004,
|
136 |
+
save_imatrix: stored collected data after 136 chunks in internlm2_5-7b-chat-IMat-GGUF/imatrix.dat
|
137 |
+
|
138 |
+
llama_print_timings: load time = 3258.43 ms
|
139 |
+
llama_print_timings: sample time = 0.00 ms / 1 runs ( 0.00 ms per token, inf tokens per second)
|
140 |
+
llama_print_timings: prompt eval time = 78728.65 ms / 69632 tokens ( 1.13 ms per token, 884.46 tokens per second)
|
141 |
+
llama_print_timings: eval time = 0.00 ms / 1 runs ( 0.00 ms per token, inf tokens per second)
|
142 |
+
llama_print_timings: total time = 81165.19 ms / 69633 tokens
|
143 |
+
|
144 |
+
Final estimate: PPL = 7.8004 +/- 0.10722
|