Eval bug: Qwerky QwQ 32B (rwkv6qwen2) failed to load #12662
Description
Name and Version
version: 5002 (2c3f8b8)
built with MSVC 19.29.30158.0 for
Operating systems
Windows
GGML backends
CUDA
Hardware
Ryzen 5700X + RTX 3090
Models
featherless-ai/Qwerky-QwQ-32B
tested with DevQuasar/featherless-ai.Qwerky-QwQ-32B-GGUF Q4_K_S
and IQ4_XS created with convert_hf_to_gguf.py locally
Problem description & steps to reproduce
launching the server/cli fails because of llama_model_load: error loading model: check_tensor_dims: tensor 'blk.0.time_mix_w1.weight' has wrong shape; expected 5120, 320, got 5120, 640, 1, 1
I tried it with the Q4_K_S model from the hub and with a locally created IQ4_XS model. Model conversion works without any errors, but loading the model fails for both quants.
First Bad Commit
No response
Relevant log output
$ llama-server.exe -m model.gguf -ctk q8_0 -ctv q8_0 --n_gpu_layers 128 --ctx-size 32768 -fa
ggml_cuda_init: GGML_CUDA_FORCE_MMQ: no
ggml_cuda_init: GGML_CUDA_FORCE_CUBLAS: no
ggml_cuda_init: found 1 CUDA devices:
Device 0: NVIDIA GeForce RTX 3090, compute capability 8.6, VMM: yes
build: 5002 (2c3f8b85) with MSVC 19.29.30158.0 for
system info: n_threads = 8, n_threads_batch = 8, total_threads = 16
system_info: n_threads = 8 (n_threads_batch = 8) / 16 | CUDA : ARCHS = 500,610,700,750,800 | USE_GRAPHS = 1 | PEER_MAX_BATCH_SIZE = 128 | CPU : SSE3 = 1 | SSSE3 = 1 | AVX = 1 | AVX2 = 1 | F16C = 1 | FMA = 1 | BMI2 = 1 | LLAMAFILE = 1 | OPENMP = 1 | AARCH64_REPACK = 1 |
main: binding port with default address family
main: HTTP server is listening, hostname: 127.0.0.1, port: 8080, http threads: 15
main: loading model
srv load_model: loading model 'model.gguf'
llama_model_load_from_file_impl: using device CUDA0 (NVIDIA GeForce RTX 3090) - 23306 MiB free
llama_model_loader: loaded meta data with 29 key-value pairs and 1283 tensors from model.gguf (version GGUF V3 (latest))
llama_model_loader: Dumping metadata keys/values. Note: KV overrides do not apply in this output.
llama_model_loader: - kv 0: general.architecture str = rwkv6qwen2
llama_model_loader: - kv 1: general.type str = model
llama_model_loader: - kv 2: general.name str = Featherless ai.Qwerky QwQ 32B
llama_model_loader: - kv 3: general.basename str = featherless-ai.Qwerky-QwQ
llama_model_loader: - kv 4: general.size_label str = 32B
llama_model_loader: - kv 5: general.license str = apache-2.0
llama_model_loader: - kv 6: rwkv6qwen2.context_length u32 = 1048576
llama_model_loader: - kv 7: rwkv6qwen2.embedding_length u32 = 5120
llama_model_loader: - kv 8: rwkv6qwen2.block_count u32 = 64
llama_model_loader: - kv 9: rwkv6qwen2.wkv.head_size u32 = 128
llama_model_loader: - kv 10: rwkv6qwen2.time_mix_extra_dim u32 = 64
llama_model_loader: - kv 11: rwkv6qwen2.time_decay_extra_dim u32 = 128
llama_model_loader: - kv 12: rwkv6qwen2.feed_forward_length u32 = 27648
llama_model_loader: - kv 13: rwkv6qwen2.attention.layer_norm_rms_epsilon f32 = 0.000010
llama_model_loader: - kv 14: rwkv6qwen2.token_shift_count u32 = 1
llama_model_loader: - kv 15: rwkv6qwen2.attention.head_count_kv u32 = 8
llama_model_loader: - kv 16: rwkv6qwen2.attention.head_count u32 = 0
llama_model_loader: - kv 17: tokenizer.ggml.model str = gpt2
llama_model_loader: - kv 18: tokenizer.ggml.pre str = qwen2
llama_model_loader: - kv 19: tokenizer.ggml.tokens arr[str,152064] = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv 20: tokenizer.ggml.token_type arr[i32,152064] = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv 21: tokenizer.ggml.merges arr[str,151387] = ["─á ─á", "─á─á ─á─á", "i n", "─á t",...
llama_model_loader: - kv 22: tokenizer.ggml.eos_token_id u32 = 151645
llama_model_loader: - kv 23: tokenizer.ggml.padding_token_id u32 = 151643
llama_model_loader: - kv 24: tokenizer.ggml.bos_token_id u32 = 151643
llama_model_loader: - kv 25: tokenizer.ggml.add_bos_token bool = false
llama_model_loader: - kv 26: tokenizer.chat_template str = {%- if tools %}\n {{- '<|im_start|>...
llama_model_loader: - kv 27: general.quantization_version u32 = 2
llama_model_loader: - kv 28: general.file_type u32 = 14
llama_model_loader: - type f32: 769 tensors
llama_model_loader: - type q4_K: 505 tensors
llama_model_loader: - type q5_K: 8 tensors
llama_model_loader: - type q6_K: 1 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type = Q4_K - Small
print_info: file size = 20.25 GiB (4.98 BPW)
load: special tokens cache size = 26
load: token to piece cache size = 0.9311 MB
print_info: arch = rwkv6qwen2
print_info: vocab_only = 0
print_info: n_ctx_train = 1048576
print_info: n_embd = 5120
print_info: n_layer = 64
print_info: n_head = 0
print_info: n_head_kv = 8
print_info: n_rot = 0
print_info: n_swa = 0
print_info: n_swa_pattern = 1
print_info: n_embd_head_k = 0
print_info: n_embd_head_v = 0
print_info: n_gqa = 0
print_info: n_embd_k_gqa = 0
print_info: n_embd_v_gqa = 0
print_info: f_norm_eps = 0.0e+00
print_info: f_norm_rms_eps = 1.0e-05
print_info: f_clamp_kqv = 0.0e+00
print_info: f_max_alibi_bias = 0.0e+00
print_info: f_logit_scale = 0.0e+00
print_info: f_attn_scale = 0.0e+00
print_info: n_ff = 27648
print_info: n_expert = 0
print_info: n_expert_used = 0
print_info: causal attn = 1
print_info: pooling type = 0
print_info: rope type = -1
print_info: rope scaling = linear
print_info: freq_base_train = 10000.0
print_info: freq_scale_train = 1
print_info: n_ctx_orig_yarn = 1048576
print_info: rope_finetuned = unknown
print_info: ssm_d_conv = 0
print_info: ssm_d_inner = 0
print_info: ssm_d_state = 0
print_info: ssm_dt_rank = 0
print_info: ssm_dt_b_c_rms = 0
print_info: model type = 32B
print_info: model params = 34.95 B
print_info: general.name = Featherless ai.Qwerky QwQ 32B
print_info: vocab type = BPE
print_info: n_vocab = 152064
print_info: n_merges = 151387
print_info: BOS token = 151643 '<|endoftext|>'
print_info: EOS token = 151645 '<|im_end|>'
print_info: EOT token = 151645 '<|im_end|>'
print_info: PAD token = 151643 '<|endoftext|>'
print_info: LF token = 198 '─è'
print_info: FIM PRE token = 151659 '<|fim_prefix|>'
print_info: FIM SUF token = 151661 '<|fim_suffix|>'
print_info: FIM MID token = 151660 '<|fim_middle|>'
print_info: FIM PAD token = 151662 '<|fim_pad|>'
print_info: FIM REP token = 151663 '<|repo_name|>'
print_info: FIM SEP token = 151664 '<|file_sep|>'
print_info: EOG token = 151643 '<|endoftext|>'
print_info: EOG token = 151645 '<|im_end|>'
print_info: EOG token = 151662 '<|fim_pad|>'
print_info: EOG token = 151663 '<|repo_name|>'
print_info: EOG token = 151664 '<|file_sep|>'
print_info: max token length = 256
load_tensors: loading model tensors, this can take a while... (mmap = true)
llama_model_load: error loading model: check_tensor_dims: tensor 'blk.0.time_mix_w1.weight' has wrong shape; expected 5120, 320, got 5120, 640, 1, 1
llama_model_load_from_file_impl: failed to load model
common_init_from_params: failed to load model 'model.gguf'