github-starred/ollama
2
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-05-07 00:22:43 -05:00
Code Issues 14.5k Packages Projects Releases 120 Wiki Activity

[GH-ISSUE #10759] --quantize leads to incorrect KVs / tokenizer #69127

New Issue
Open
opened 2026-05-04 17:14:19 -05:00 by GiteaMirror · 0 comments
GiteaMirror commented 2026-05-04 17:14:19 -05:00
Owner
Copy Link

Originally created by @jmorganca on GitHub (May 17, 2025).
Original GitHub issue: https://github.com/ollama/ollama/issues/10759

What is the issue?

llama_model_load_from_file_impl: using device Metal (Apple M3 Max) - 98303 MiB free
llama_model_loader: loaded meta data with 29 key-value pairs and 363 tensors from /Users/jmorgan/.ollama/models/blobs/sha256-f05ccecc2baeb1d3c4c68b0a61c7b17124fd04efdb8f80fdbb10d0c15df8d8f1 (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.file_type u32              = 15
llama_model_loader: - kv   2:                    general.parameter_count u64              = 23572403200
llama_model_loader: - kv   3:               general.quantization_version u32              = 2
llama_model_loader: - kv   4:                         general.size_label str              = 24B
llama_model_loader: - kv   5:                               general.type str              = model
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.attention.key_length u32              = 128
llama_model_loader: - kv   9:     llama.attention.layer_norm_rms_epsilon f32              = 0.000010
llama_model_loader: - kv  10:               llama.attention.value_length u32              = 128
llama_model_loader: - kv  11:                          llama.block_count u32              = 40
llama_model_loader: - kv  12:                       llama.context_length u32              = 32768
llama_model_loader: - kv  13:                     llama.embedding_length u32              = 5120
llama_model_loader: - kv  14:                  llama.feed_forward_length u32              = 32768
llama_model_loader: - kv  15:                 llama.rope.dimension_count u32              = 128
llama_model_loader: - kv  16:                       llama.rope.freq_base f32              = 1000000000.000000
llama_model_loader: - kv  17:                           llama.vocab_size u32              = 131072
llama_model_loader: - kv  18:               tokenizer.ggml.add_bos_token bool             = true
llama_model_loader: - kv  19:               tokenizer.ggml.add_eos_token bool             = false
llama_model_loader: - kv  20:            tokenizer.ggml.add_space_prefix bool             = false
llama_model_loader: - kv  21:                tokenizer.ggml.bos_token_id u32              = 1
llama_model_loader: - kv  22:                tokenizer.ggml.eos_token_id u32              = 2
llama_model_loader: - kv  23:                      tokenizer.ggml.merges arr[str,0]       = []
llama_model_loader: - kv  24:                       tokenizer.ggml.model str              = gpt2
llama_model_loader: - kv  25:                         tokenizer.ggml.pre str              = tekken
llama_model_loader: - kv  26:                  tokenizer.ggml.token_type arr[i32,0]       = []
llama_model_loader: - kv  27:                      tokenizer.ggml.tokens arr[str,0]       = []
llama_model_loader: - kv  28:            tokenizer.ggml.unknown_token_id u32              = 0
llama_model_loader: - type  f32:   81 tensors
llama_model_loader: - type q4_K:  241 tensors
llama_model_loader: - type q6_K:   41 tensors
print_info: file format = GGUF V3 (latest)
print_info: file type   = Q4_K - Medium
print_info: file size   = 13.34 GiB (4.86 BPW) 
load: model vocab missing newline token, using special_pad_id instead
load: bad special token: 'tokenizer.ggml.bos_token_id' = 1, using default id 11
load: bad special token: 'tokenizer.ggml.eos_token_id' = 2, using default id 11
load: bad special token: 'tokenizer.ggml.unknown_token_id' = 0, using default id -1
load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect
load: special tokens cache size = 0
load: token to piece cache size = 0.0000 MB

Relevant log output

OS

No response

GPU

No response

CPU

No response

Ollama version

No response

Originally created by @jmorganca on GitHub (May 17, 2025). Original GitHub issue: https://github.com/ollama/ollama/issues/10759 ### What is the issue? ``` llama_model_load_from_file_impl: using device Metal (Apple M3 Max) - 98303 MiB free llama_model_loader: loaded meta data with 29 key-value pairs and 363 tensors from /Users/jmorgan/.ollama/models/blobs/sha256-f05ccecc2baeb1d3c4c68b0a61c7b17124fd04efdb8f80fdbb10d0c15df8d8f1 (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.file_type u32 = 15 llama_model_loader: - kv 2: general.parameter_count u64 = 23572403200 llama_model_loader: - kv 3: general.quantization_version u32 = 2 llama_model_loader: - kv 4: general.size_label str = 24B llama_model_loader: - kv 5: general.type str = model 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.attention.key_length u32 = 128 llama_model_loader: - kv 9: llama.attention.layer_norm_rms_epsilon f32 = 0.000010 llama_model_loader: - kv 10: llama.attention.value_length u32 = 128 llama_model_loader: - kv 11: llama.block_count u32 = 40 llama_model_loader: - kv 12: llama.context_length u32 = 32768 llama_model_loader: - kv 13: llama.embedding_length u32 = 5120 llama_model_loader: - kv 14: llama.feed_forward_length u32 = 32768 llama_model_loader: - kv 15: llama.rope.dimension_count u32 = 128 llama_model_loader: - kv 16: llama.rope.freq_base f32 = 1000000000.000000 llama_model_loader: - kv 17: llama.vocab_size u32 = 131072 llama_model_loader: - kv 18: tokenizer.ggml.add_bos_token bool = true llama_model_loader: - kv 19: tokenizer.ggml.add_eos_token bool = false llama_model_loader: - kv 20: tokenizer.ggml.add_space_prefix bool = false llama_model_loader: - kv 21: tokenizer.ggml.bos_token_id u32 = 1 llama_model_loader: - kv 22: tokenizer.ggml.eos_token_id u32 = 2 llama_model_loader: - kv 23: tokenizer.ggml.merges arr[str,0] = [] llama_model_loader: - kv 24: tokenizer.ggml.model str = gpt2 llama_model_loader: - kv 25: tokenizer.ggml.pre str = tekken llama_model_loader: - kv 26: tokenizer.ggml.token_type arr[i32,0] = [] llama_model_loader: - kv 27: tokenizer.ggml.tokens arr[str,0] = [] llama_model_loader: - kv 28: tokenizer.ggml.unknown_token_id u32 = 0 llama_model_loader: - type f32: 81 tensors llama_model_loader: - type q4_K: 241 tensors llama_model_loader: - type q6_K: 41 tensors print_info: file format = GGUF V3 (latest) print_info: file type = Q4_K - Medium print_info: file size = 13.34 GiB (4.86 BPW) load: model vocab missing newline token, using special_pad_id instead load: bad special token: 'tokenizer.ggml.bos_token_id' = 1, using default id 11 load: bad special token: 'tokenizer.ggml.eos_token_id' = 2, using default id 11 load: bad special token: 'tokenizer.ggml.unknown_token_id' = 0, using default id -1 load: special_eos_id is not in special_eog_ids - the tokenizer config may be incorrect load: special tokens cache size = 0 load: token to piece cache size = 0.0000 MB ``` ### Relevant log output ```shell ``` ### OS _No response_ ### GPU _No response_ ### CPU _No response_ ### Ollama version _No response_
GiteaMirror added the bug label 2026-05-04 17:14:19 -05:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
amd
api
app
bug
build
cli
cloud
compatibility
context-length
create
docker
documentation
embeddings
feature request
feedback wanted
good first issue
gpt-oss
gpu
harmony
help wanted
image
install
intel
js
launch
linux
macos
memory
mlx
model
needs more info
networking
nvidia
ollama.com
performance
pull-request
Mirrored from GitHub Pull Request
 python
question
registry
rendering
thinking
tools
top
vulkan
windows
wsl
No Label bug
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
GiteaMirror ninjasurge
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: github-starred/ollama#69127
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?