finetune error: ggml_flash_attn_ext() not yet supported

[michaelmidura]

I have been finetuning a model based on Meta-Llama-3-8B using finetune. The model was downloaded from the meta-llama Hugging Face. I am running macOS on Apple Silicon. I recently updated llama.cpp to b2989 (27891f6) which broke finetune.

Steps to reproduce:

1. make clean and make
2. Convert model using python convert-hf-to-gguf.py models/Meta-Llama-3-8B/
3. Finetune using ./finetune --model-base models/Meta-Llama-3-8B/ggml-model-f16.gguf --lora-out lora-test-0x00001.bin --train-data shakespeare.txt --threads 6 --adam-iter 30 --batch 4 --ctx 64 --save-every 10 --use-checkpointing

Output:

main: seed: 1716580115
main: model base = 'models/Meta-Llama-3-8B/ggml-model-f16.gguf'
llama_model_loader: loaded meta data with 21 key-value pairs and 291 tensors from models/Meta-Llama-3-8B/ggml-model-f16.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              = llama
llama_model_loader: - kv   1:                               general.name str              = Meta-Llama-3-8B
llama_model_loader: - kv   2:                          llama.block_count u32              = 32
llama_model_loader: - kv   3:                       llama.context_length u32              = 8192
llama_model_loader: - kv   4:                     llama.embedding_length u32              = 4096
llama_model_loader: - kv   5:                  llama.feed_forward_length u32              = 14336
llama_model_loader: - kv   6:                 llama.attention.head_count u32              = 32
llama_model_loader: - kv   7:              llama.attention.head_count_kv u32              = 8
llama_model_loader: - kv   8:                       llama.rope.freq_base f32              = 500000.000000
llama_model_loader: - kv   9:     llama.attention.layer_norm_rms_epsilon f32              = 0.000010
llama_model_loader: - kv  10:                          general.file_type u32              = 1
llama_model_loader: - kv  11:                           llama.vocab_size u32              = 128256
llama_model_loader: - kv  12:                 llama.rope.dimension_count u32              = 128
llama_model_loader: - kv  13:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  14:                         tokenizer.ggml.pre str              = llama-bpe
llama_model_loader: - kv  15:                      tokenizer.ggml.tokens arr[str,128256]  = ["!", "\"", "#", "$", "%", "&", "'", ...
llama_model_loader: - kv  16:                  tokenizer.ggml.token_type arr[i32,128256]  = [1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, ...
llama_model_loader: - kv  17:                      tokenizer.ggml.merges arr[str,280147]  = ["Ġ Ġ", "Ġ ĠĠĠ", "ĠĠ ĠĠ", "...
llama_model_loader: - kv  18:                tokenizer.ggml.bos_token_id u32              = 128000
llama_model_loader: - kv  19:                tokenizer.ggml.eos_token_id u32              = 128001
llama_model_loader: - kv  20:               general.quantization_version u32              = 2
llama_model_loader: - type  f32:   65 tensors
llama_model_loader: - type  f16:  226 tensors
llm_load_vocab: special tokens definition check successful ( 256/128256 ).
llm_load_print_meta: format           = GGUF V3 (latest)
llm_load_print_meta: arch             = llama
llm_load_print_meta: vocab type       = BPE
llm_load_print_meta: n_vocab          = 128256
llm_load_print_meta: n_merges         = 280147
llm_load_print_meta: n_ctx_train      = 8192
llm_load_print_meta: n_embd           = 4096
llm_load_print_meta: n_head           = 32
llm_load_print_meta: n_head_kv        = 8
llm_load_print_meta: n_layer          = 32
llm_load_print_meta: n_rot            = 128
llm_load_print_meta: n_embd_head_k    = 128
llm_load_print_meta: n_embd_head_v    = 128
llm_load_print_meta: n_gqa            = 4
llm_load_print_meta: n_embd_k_gqa     = 1024
llm_load_print_meta: n_embd_v_gqa     = 1024
llm_load_print_meta: f_norm_eps       = 0.0e+00
llm_load_print_meta: f_norm_rms_eps   = 1.0e-05
llm_load_print_meta: f_clamp_kqv      = 0.0e+00
llm_load_print_meta: f_max_alibi_bias = 0.0e+00
llm_load_print_meta: f_logit_scale    = 0.0e+00
llm_load_print_meta: n_ff             = 14336
llm_load_print_meta: n_expert         = 0
llm_load_print_meta: n_expert_used    = 0
llm_load_print_meta: causal attn      = 1
llm_load_print_meta: pooling type     = 0
llm_load_print_meta: rope type        = 0
llm_load_print_meta: rope scaling     = linear
llm_load_print_meta: freq_base_train  = 500000.0
llm_load_print_meta: freq_scale_train = 1
llm_load_print_meta: n_yarn_orig_ctx  = 8192
llm_load_print_meta: rope_finetuned   = unknown
llm_load_print_meta: ssm_d_conv       = 0
llm_load_print_meta: ssm_d_inner      = 0
llm_load_print_meta: ssm_d_state      = 0
llm_load_print_meta: ssm_dt_rank      = 0
llm_load_print_meta: model type       = 8B
llm_load_print_meta: model ftype      = F16
llm_load_print_meta: model params     = 8.03 B
llm_load_print_meta: model size       = 14.96 GiB (16.00 BPW)
llm_load_print_meta: general.name     = Meta-Llama-3-8B
llm_load_print_meta: BOS token        = 128000 '<|begin_of_text|>'
llm_load_print_meta: EOS token        = 128001 '<|end_of_text|>'
llm_load_print_meta: LF token         = 128 'Ä'
llm_load_print_meta: EOT token        = 128009 '<|eot_id|>'
llm_load_tensors: ggml ctx size =    0.15 MiB
llm_load_tensors: offloading 0 repeating layers to GPU
llm_load_tensors: offloaded 0/33 layers to GPU
llm_load_tensors:        CPU buffer size = 15317.02 MiB
.........................................................................................
llama_new_context_with_model: n_ctx      = 512
llama_new_context_with_model: n_batch    = 512
llama_new_context_with_model: n_ubatch   = 512
llama_new_context_with_model: flash_attn = 0
llama_new_context_with_model: freq_base  = 500000.0
llama_new_context_with_model: freq_scale = 1
llama_kv_cache_init:        CPU KV buffer size =    64.00 MiB
llama_new_context_with_model: KV self size  =   64.00 MiB, K (f16):   32.00 MiB, V (f16):   32.00 MiB
llama_new_context_with_model:        CPU  output buffer size =     0.49 MiB
llama_new_context_with_model:        CPU compute buffer size =   258.50 MiB
llama_new_context_with_model: graph nodes  = 1030
llama_new_context_with_model: graph splits = 1
main: init model
print_params: n_vocab               : 128256
print_params: n_ctx                 : 64
print_params: n_embd                : 4096
print_params: n_ff                  : 14336
print_params: n_head                : 32
print_params: n_head_kv             : 8
print_params: n_layer               : 32
print_params: norm_rms_eps          : 0.000010
print_params: rope_freq_base        : 500000.000000
print_params: rope_freq_scale       : 1.000000
print_lora_params: n_rank_attention_norm : 1
print_lora_params: n_rank_wq             : 4
print_lora_params: n_rank_wk             : 4
print_lora_params: n_rank_wv             : 4
print_lora_params: n_rank_wo             : 4
print_lora_params: n_rank_ffn_norm       : 1
print_lora_params: n_rank_ffn_gate       : 4
print_lora_params: n_rank_ffn_down       : 4
print_lora_params: n_rank_ffn_up         : 4
print_lora_params: n_rank_tok_embeddings : 4
print_lora_params: n_rank_norm           : 1
print_lora_params: n_rank_output         : 4
main: total train_iterations 0
main: seen train_samples     0
main: seen train_tokens      0
main: completed train_epochs 0
main: lora_size = 94956320 bytes (90.6 MB)
main: opt_size  = 141731824 bytes (135.2 MB)
main: opt iter 0
main: input_size = 131335200 bytes (125.3 MB)
GGML_ASSERT: examples/finetune/finetune.cpp:646: false && "TODO: ggml_flash_attn_ext() not yet supported"
fish: Job 1, './finetune --model-base models/…' terminated by signal SIGABRT (Abort)

[michaelmidura]

finetune works on b2974 (1b1e27c)
finetune broken after b2976 (d48c88c)

[ggerganov]

Should be fixed with 9588f19

[amlan-sw]

finetune works on b2974 (1b1e27c) finetune broken after b2976 (d48c88c)

same error confirmed,
model: Lexi-Llama-3-8B-Uncensored_Q4_K_M.gguf
work : llama-b2974-bin-win-vulkan-x64.zip
broken-tested after 2974 : 2984,3008

[ggerganov]

Don't use --use-flash in the command line args

[amlan-sw]

Don't use --use-flash in the command line args

thx, 3030 run USING : --no-flash

[github-actions]

This issue was closed because it has been inactive for 14 days since being marked as stale.