Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
295 commits
Select commit Hold shift + click to select a range
c5bf27f
All 3 metrics now exceed Q3_K_M
GeoffApples Dec 13, 2025
1cf26dc
Documentation updated
GeoffApples Dec 13, 2025
9b58d82
Merge pull request #4 from GeoffApples/Q3_HIFI_1.7B_fast
geoffmunn Dec 13, 2025
0baa2c8
Q3_HIFI_A now the official version
GeoffApples Dec 13, 2025
bc8ba8a
Merge pull request #5 from GeoffApples/Q3_HIFI_1.7B_fast
geoffmunn Dec 13, 2025
2d4d0b3
Speed benchmark script added
GeoffApples Dec 14, 2025
a177f2c
Merge pull request #6 from GeoffApples/Q3_HIFI_1.7B_fast
geoffmunn Dec 14, 2025
bc3c5cf
Merge pull request #7 from ggml-org/master
geoffmunn Dec 14, 2025
0e6f3aa
Merge branch 'Q3_HIFI' into master
geoffmunn Dec 14, 2025
9971857
Merge pull request #8 from geoffmunn/master
geoffmunn Dec 14, 2025
42b6477
Old files removed
Dec 21, 2025
5792ab4
Cross-model documentation added
Dec 21, 2025
8b72146
Validation errors fixed
Dec 21, 2025
daf0e20
Whitespace fixed
Dec 21, 2025
bf0d021
Whitespace fixes
Dec 21, 2025
f79424e
Whitespace fixes
Dec 21, 2025
abcb4cc
Whitespace fixes
Dec 21, 2025
7724f7b
Whitespace changes
Dec 21, 2025
a6bb077
Whitespace fixes
Dec 21, 2025
9bae334
Whitespace fixes
Dec 21, 2025
dce3e67
Whitespace fixes
Dec 21, 2025
3e3f931
Whitespace fixes
Dec 21, 2025
972d662
Whitespace fixes
Dec 21, 2025
20390e2
Whitespace fixes
Dec 21, 2025
4851a00
print statements changed to logging()
Dec 21, 2025
9be1c3d
Extra blank line removed
Dec 21, 2025
c42d48f
Merge pull request #9 from geoffmunn/Q3_HIFI
geoffmunn Dec 21, 2025
dbf9a9a
Documentation moved
Dec 21, 2025
2c4049e
GGML_TYPE_Q3_HIFI now value 12
Dec 21, 2025
e4fd98f
GGML_TYPE_Q3_HIFI moved to end, numbers re-ordered
Dec 21, 2025
beb72af
Missing files added
GeoffApples Dec 27, 2025
c553255
Added new ftype enum
GeoffApples Dec 27, 2025
73dd524
Added tensor type selection logic
GeoffApples Dec 27, 2025
d44e778
Added CLI entry
GeoffApples Dec 27, 2025
b491e3e
Added description string
GeoffApples Dec 27, 2025
325d34e
Phase 2 improvements
GeoffApples Dec 27, 2025
62acf48
V3 changes
GeoffApples Dec 27, 2025
94b18a1
Build error fixed
GeoffApples Dec 27, 2025
969754f
Build warning fixed
GeoffApples Dec 27, 2025
30535c6
Missing type added
GeoffApples Dec 27, 2025
7c233ec
Quantisation error fixed
GeoffApples Dec 27, 2025
e7862d1
Outlier budget and early exit
GeoffApples Dec 27, 2025
13e8b25
Build error fixed
GeoffApples Dec 27, 2025
7249d24
Missing type added
GeoffApples Dec 27, 2025
2128f33
Q4_HIFI now standardised
GeoffApples Dec 27, 2025
e6ab6f6
INT8 residuals for size reduction
GeoffApples Dec 30, 2025
8543275
Unused variables removed
GeoffApples Dec 30, 2025
e443866
Add HIFI quantization support with layer-adaptive outlier allocation
GeoffApples Dec 31, 2025
131c0e7
Refactor thread-local storage declaration and improve tensor importan…
GeoffApples Dec 31, 2025
aaa3564
Add maximum outliers definition for Q6_K_HIFI_RES8 format
GeoffApples Dec 31, 2025
5411916
Update include path for HIFI quantization header
GeoffApples Dec 31, 2025
0e0830a
Enhance model parameter calculation and logging for HIFI quantization
GeoffApples Dec 31, 2025
8077951
Add model size parameter to HIFI quantization context
GeoffApples Dec 31, 2025
3b40c30
Enhance layer-adaptive outlier allocation for Q4_HIFI quantization
GeoffApples Dec 31, 2025
344495f
Parameter finetuning
GeoffApples Dec 31, 2025
1762085
Refine scale-dependent adjustments for outlier allocation in HIFI qua…
GeoffApples Dec 31, 2025
e9a5e7c
Refine scale-dependent adjustments for outlier allocation in HIFI qua…
GeoffApples Dec 31, 2025
a246d6e
Refine model parameter calculation and scale adjustments for HIFI qua…
GeoffApples Dec 31, 2025
eed04a7
Missing constants added
GeoffApples Jan 1, 2026
79e1751
Implement Q6_K_HIFI_RES8 kernel with residual corrections
GeoffApples Jan 1, 2026
0ed8ad0
Test to see what is happening in the GPU implementation
GeoffApples Jan 1, 2026
581f3ea
First round of size reductions
GeoffApples Jan 1, 2026
e30d855
Option2 of size reductions
GeoffApples Jan 1, 2026
b1312f3
Merge pull request #10 from GeoffApples/Q4_HIFI_v3
geoffmunn Jan 1, 2026
c29a18f
Add quantization type string for Hugging Face model card display
GeoffApples Jan 1, 2026
c4afeb9
Merge pull request #12 from GeoffApples/Q4_HIFI_v3
geoffmunn Jan 3, 2026
3ccfcd3
Q4_HIFI renamed to Q4_K_HIFI
GeoffApples Jan 4, 2026
48e01fb
Add Q5_K_HIFI_RES8 quantization format and associated functions
GeoffApples Jan 4, 2026
9127308
Update Q5_K_HIFI_RES8 structure size and padding initialization
GeoffApples Jan 4, 2026
1782b40
Enhance Q5_K_HIFI_RES8 dequantization and dot product functions
GeoffApples Jan 4, 2026
a6d58d7
Refactor Q5_K_HIFI_RES8 quantization function names for consistency
GeoffApples Jan 4, 2026
339080d
Enhance Q5_K_HIFI_RES8 quantization support in CPU operations
GeoffApples Jan 4, 2026
4c9a074
Add maximum outliers definition for Q5_K_HIFI_RES8 format
GeoffApples Jan 4, 2026
ac65290
Refactor Q5_K_HIFI_RES8 quantization function for improved clarity
GeoffApples Jan 5, 2026
909fa27
Build warnings fixed
GeoffApples Jan 5, 2026
8b2338d
2 extra strategies implemented
GeoffApples Jan 5, 2026
e5c0c28
Merge pull request #13 from GeoffApples/Q4_K_HIFI
geoffmunn Jan 5, 2026
ec73ae6
Improvements for small (0.6B) models
GeoffApples Jan 5, 2026
5d1f6dc
Merge pull request #14 from GeoffApples/Q4_K_HIFI
geoffmunn Jan 5, 2026
83fc186
Add Q5_K_HIFI_HYBRID quantization format for small models
GeoffApples Jan 5, 2026
0e26b4b
Merge pull request #15 from GeoffApples/Q4_K_HIFI
geoffmunn Jan 5, 2026
67e4f38
Build errors fixed
GeoffApples Jan 5, 2026
340cb56
Merge pull request #16 from GeoffApples/Q4_K_HIFI
geoffmunn Jan 5, 2026
1c40df1
Missing type added
GeoffApples Jan 5, 2026
828a0cc
Merge pull request #17 from GeoffApples/Q4_K_HIFI
geoffmunn Jan 5, 2026
d44aa64
Reverted back to RES8 approach
GeoffApples Jan 5, 2026
8fb070e
Merge pull request #18 from GeoffApples/Q4_K_HIFI
geoffmunn Jan 5, 2026
32c3619
Update enhancement threshold for medium-large models in llama-quant.cpp
GeoffApples Jan 6, 2026
cb99f26
Merge pull request #19 from GeoffApples/Q4_K_HIFI
geoffmunn Jan 6, 2026
2239f3f
Refine HIFI type selection based on model size in llama-quant.cpp
GeoffApples Jan 6, 2026
a8b65cb
Merge pull request #20 from GeoffApples/Q4_K_HIFI
geoffmunn Jan 6, 2026
d1ccc0f
Test to see if 14B can be made better
GeoffApples Jan 6, 2026
7fa8e76
Merge pull request #21 from GeoffApples/Q4_K_HIFI
geoffmunn Jan 6, 2026
7fdb40c
q3_hifi renamed
GeoffApples Jan 8, 2026
c55c8af
Q3_K_HIFI added
GeoffApples Jan 9, 2026
931dd7b
Q3_K_HIFI added
GeoffApples Jan 9, 2026
873d28d
Memory usage added to benchmark script
GeoffApples Jan 9, 2026
39718b6
Display issues fixed
GeoffApples Jan 9, 2026
19e75a8
Enhance benchmark script with new models and memory tracking
geoffmunn Jan 9, 2026
02c1235
Merge branch 'master' into Q3_K_HIFI_v3
GeoffApples Jan 10, 2026
208f737
Merge pull request #22 from GeoffApples/Q3_K_HIFI_v3
geoffmunn Jan 10, 2026
5656236
Add Baseline model to benchmark speed test
geoffmunn Jan 10, 2026
3dd54cb
Add Q5_K_HIFI quantization type and enhance llama tensor handling.
GeoffApples Jan 10, 2026
c6938df
Fix typo in outlier constant name in dequantize_row_q3_k_hifi function
Jan 10, 2026
2fef475
Suppress false positive buffer overflow warnings in make_block_q4_0x4…
Jan 10, 2026
b770e74
Update outlier extension size in block_q3_k_hifi and enhance error lo…
Jan 10, 2026
3a9ede1
Add memory layout validation for Q5_K_HIFI and Q6_K_HIFI structures
GeoffApples Jan 13, 2026
e574ffb
Refactor memory layout validation for Q5_K_HIFI and Q6_K_HIFI structu…
GeoffApples Jan 13, 2026
f78c612
Merge pull request #23 from GeoffApples/Q5_K_HIFI
geoffmunn Jan 14, 2026
7e76ceb
Merge pull request #1 from geoffmunn/master
GeoffApples Jan 15, 2026
b6dc77c
Add Q3_K_HIFI adaptive enhancement functions for scale-aware tensor s…
GeoffApples Jan 15, 2026
501269a
Refine Q3_HIFI enhancement thresholds for model size categories in gg…
GeoffApples Jan 16, 2026
3c1480f
Q3_K_HIFI rescue plan for 4B models
GeoffApples Jan 17, 2026
a557bf7
Enhance Q3_K_HIFI upgrade logic based on model size categories. Clari…
GeoffApples Jan 17, 2026
893e8ab
Fixes for Apple Silicon build support
Jan 17, 2026
b1302bb
Update dequantization functions in Q3_K_HIFI to use float instead of …
Jan 17, 2026
54fb6ea
Refactor dequantization logic in Q3_K_HIFI to simplify processing and…
Jan 17, 2026
9d49955
Add support for Q5_K_HIFI and Q6_K_HIFI quantization types in Metal d…
Jan 17, 2026
2732713
Refactor kernel name generation in Metal device pipeline to utilize H…
Jan 17, 2026
65b2d66
Reintroduce HIFI type mapping for kernel name generation in Metal dev…
Jan 17, 2026
f844bc6
Merge remote-tracking branch 'upstream/master'
Jan 17, 2026
4660add
Add Q3_K_HIFI_RES8 variant for lean INT8 residual quantization. Imple…
GeoffApples Jan 18, 2026
720bfb0
Add validation for Q3_K_HIFI_RES8 data in ggml-quants.c. Implemented …
GeoffApples Jan 18, 2026
64a0b81
Add support for Q3_K_HIFI_RES8 in CUDA backend. Updated device operat…
GeoffApples Jan 18, 2026
8235400
Add Q3_K_HIFI_RES8 support in CPU operations. Updated multiple comput…
GeoffApples Jan 18, 2026
ad00c42
Refine Q3_K_HIFI quantization strategy based on model size and imatri…
GeoffApples Jan 18, 2026
200139f
Merge pull request #24 from GeoffApples/Q3_K_HIFI_v4
geoffmunn Jan 19, 2026
2c84cdd
Refine Q3_K_HIFI quantization logic to adaptively select tensor types…
GeoffApples Jan 20, 2026
95f7b76
Update Q3_K_HIFI quantization logic to reflect new findings on model …
GeoffApples Jan 20, 2026
0d94aef
Merge pull request #25 from GeoffApples/Q3_K_HIFI_v4
geoffmunn Jan 20, 2026
80fffbd
Merge branch 'ggml-org:master' into master
geoffmunn Jan 20, 2026
7bfdd84
Merge branch 'ggml-org:master' into master
geoffmunn Jan 22, 2026
9b5a3a3
Files not needed
Jan 25, 2026
3bcbede
Enhance Q3_K_HIFI structure and dequantization logic
Jan 25, 2026
7f3e1cc
Add support for Q3_K_HIFI quantization type in model loader.
Jan 25, 2026
096e43d
Enhance Q3_K_HIFI quantization and dequantization with debug logging
Jan 25, 2026
578b03a
Refactor Q3_K_HIFI quantization logic to always use Q3_K_HIFI type wh…
Jan 25, 2026
1746782
Update debug logging in Q3_K_HIFI quantization and dequantization fun…
Jan 25, 2026
714a07f
Refine GCC diagnostic handling in repack.cpp
Jan 25, 2026
3f9afca
Remove unnecessary preprocessor directive for Clang in repack.cpp
Jan 25, 2026
38f7051
Enhance Q3_K_HIFI tensor upgrade logic in llama_tensor_get_type function
Jan 25, 2026
7203ae4
Implement dedicated kernel name for Q3_K_HIFI and optimize outlier ha…
Jan 25, 2026
bd7bd03
Refactor outlier correction logic in kernel_mul_mv_q3_k_hifi_f32_impl…
Jan 25, 2026
6aa70b9
Refine Q3_K_HIFI tensor upgrade logic in llama_tensor_get_type function
Jan 25, 2026
fdfbaad
Refine Q3_K_HIFI upgrade conditions in llama_tensor_get_type function
Jan 25, 2026
0b47991
Enhance Q3_K_HIFI upgrade logic in llama_tensor_get_type function
Jan 25, 2026
1fb5494
Enhance Q3_K_HIFI quantization with model-size-aware outlier handling
Jan 26, 2026
e3c39f9
Enhance Q3_K_HIFI quantization with true outlier extraction
Jan 26, 2026
c9d9325
Refactor Q3_K_HIFI quantization to utilize residual corrections for o…
Jan 26, 2026
931da4e
Fix attempt for bad PPL readings
Jan 26, 2026
0044642
Refactor Q3_K_HIFI quantization to implement true outlier extraction
Jan 26, 2026
228fcf9
Refactor Q3_K_HIFI quantization to implement residual-based outlier c…
Jan 26, 2026
d107653
There is no way this will work, but it's a new change of direction
Jan 26, 2026
411acab
Build errors fixed
Jan 26, 2026
a9b311e
New SPARSE layout approach
Jan 26, 2026
61bd054
Build errors fixed
Jan 26, 2026
3981917
Bad perplexity values fixed
Jan 26, 2026
2484a78
Refactor Q3_K_HIFI quantization for sparse layout optimization
Jan 26, 2026
aed2bd5
Garbage perplexity hopefully fixed
Jan 26, 2026
7c072fa
Refactor Q3_K_HIFI quantization to improve outlier handling
Jan 26, 2026
1918e76
Build warnings fixed
Jan 26, 2026
026fa1d
Add q_offset calculation to kernel_mul_mv_q3_k_hifi_f32_impl
Jan 26, 2026
339651b
CUDA should match Metal
Feb 2, 2026
281d10e
ix Q3_K_HIFI struct alignment for CUDA builds
Feb 3, 2026
424f17d
Fix Q3_K_HIFI CUDA warnings and struct size
Feb 3, 2026
701a381
Fix Q3_K_HIFI vec_dot kernel - use simple addition instead of residua…
Feb 3, 2026
001aa08
Refactor vec_dot_q3_k_hifi_q8_1 kernel to improve outlier processing.
Feb 3, 2026
1dfd07f
New Metal fixes and CUDA updates
Feb 3, 2026
3012872
Fast and high precision
Feb 3, 2026
6cd68e7
Build errors in CUDA fixed
Feb 3, 2026
7eacdb4
More CUDA PPL errors fixed
Feb 4, 2026
aa92be8
Merge pull request #31 from geoffmunn/Q3_K_HIFI_CUDA
geoffmunn Feb 4, 2026
8f0a0c6
Refactor outlier magnitude calculations in Q3_K_HIFI to use double pr…
Feb 4, 2026
3b2aa1a
Add per-tensor outlier and importance control for Q3_K_HIFI quantization
Feb 8, 2026
9a3b610
Merge pull request #32 from geoffmunn/Q3_K_HIFI_large
geoffmunn Feb 8, 2026
8dccd86
Coder imatrix script
Feb 9, 2026
8f9a6b5
Faster version
Feb 9, 2026
ad67f49
Changes reverted
Feb 9, 2026
df357a9
Q4_K_HIFI build changes
Feb 10, 2026
e4d3fcf
Build errors fixed
Feb 10, 2026
196a0f5
Add validation for Q4_K_HIFI quantization data.
Feb 10, 2026
3079eac
Enhance IMatrixCollector to handle non-finite activations by introduc…
geoffmunn Feb 12, 2026
a13b75d
Change reverted
geoffmunn Feb 12, 2026
ee8494f
Merge pull request #33 from geoffmunn/Q4_K_HIFI_large
geoffmunn Feb 13, 2026
546da98
First phase of Q5_K_HIFI completed
geoffmunn Feb 13, 2026
2b5b708
Step B implemented
geoffmunn Feb 13, 2026
f6b163d
Build errors fixed
geoffmunn Feb 13, 2026
6c4d122
Optimize residual correction in ggml_vec_dot_q5_k_hifi_res8_q8_K by i…
geoffmunn Feb 13, 2026
3f599e6
Implement fused single-pass dequantization with residual application …
geoffmunn Feb 13, 2026
9189078
Refactor dequantization logic in ggml-quants.c to use direct access t…
geoffmunn Feb 13, 2026
d828133
Merge pull request #34 from geoffmunn/Q5_K_HIFI
geoffmunn Feb 16, 2026
ecda105
Optimize dequantization in ggml-cuda by refining residual application…
geoffmunn Feb 16, 2026
3826d72
Merge pull request #36 from geoffmunn/Q5_K_HIFI
geoffmunn Feb 16, 2026
1c19ebc
Merge remote-tracking branch 'upstream/master'
geoffmunn Feb 16, 2026
1968d8b
Refactor mul_mat_vec_q_switch_type to include ids_stride parameter fo…
geoffmunn Feb 16, 2026
1aca80f
Imatrix reverted back to latest version
geoffmunn Feb 17, 2026
54da5b4
Project specifics updated
geoffmunn Feb 19, 2026
9e97af4
Revise README for HIFI Quantisation Fork
geoffmunn Feb 19, 2026
f930b67
Merge branch 'master' of https://github.com/geoffmunn/llama.cpp
geoffmunn Feb 19, 2026
66cecd5
Add HIFI model build guide
geoffmunn Feb 20, 2026
cff2e24
Delete Q4_K_HIFI_ROADMAP.md
geoffmunn Feb 20, 2026
1c5bb7e
Q2_K_HIFI architecture implementation
geoffmunn Feb 22, 2026
0295de5
Add validation for Q2_K_HIFI quantization data in ggml-quants.c
geoffmunn Feb 22, 2026
1fc5d5a
Add Q2_K_HIFI support in Metal implementation, including dequantizati…
geoffmunn Feb 22, 2026
793dcf2
Refactor Q2_K_HIFI quantization to utilize FP16 outlier preservation …
geoffmunn Feb 22, 2026
831aced
Add Q2_K_HIFI case handling in ggml_compute_forward functions for imp…
geoffmunn Feb 23, 2026
f544ace
Add dual-mode support for Q2_K_HIFI quantization, enabling both outli…
geoffmunn Feb 23, 2026
3af0f1a
Refine Q2_K_HIFI enhancement logic in llama_tensor_get_type function.…
geoffmunn Feb 23, 2026
96d1ca6
CUDA and Vulkan support added
geoffmunn Feb 24, 2026
171f053
renamed the local n to n_corr
geoffmunn Feb 24, 2026
904640a
First cut at speed increases
geoffmunn Feb 27, 2026
5e957b9
First cut at speed increases
geoffmunn Feb 27, 2026
3f8d792
Merge branch 'Q5_K_HIFI' of https://github.com/geoffmunn/llama.cpp in…
geoffmunn Feb 27, 2026
484d954
Improved speed attempt
geoffmunn Feb 27, 2026
423f87b
Better error messages
geoffmunn Feb 28, 2026
dad4831
Requirements added
geoffmunn Feb 28, 2026
d6abc0a
Merge branch 'master' of https://github.com/geoffmunn/llama.cpp
geoffmunn Feb 28, 2026
1d01115
Merge branch 'master' of https://github.com/ggml-org/llama.cpp
geoffmunn Feb 28, 2026
6856a7f
Metal kernal improvements
geoffmunn Mar 1, 2026
4bc13c2
Build error fixed
geoffmunn Mar 1, 2026
3252ed2
NaN errors fixed
geoffmunn Mar 1, 2026
c4861e9
Add support for Q5_K_HIFI_RES8 layout in mmq_get_q8_1_ds_layout function
geoffmunn Mar 2, 2026
1a6936d
Merge pull request #38 from geoffmunn/Q2_K_HIFI_v2
geoffmunn Mar 2, 2026
d74f464
Merge pull request #39 from geoffmunn/Q5_K_HIFI
geoffmunn Mar 2, 2026
48a3b7f
Whitespace fixed
geoffmunn Mar 2, 2026
a1a2687
Whitespace fixed
geoffmunn Mar 2, 2026
74d62d2
Whitespace fixes for linter
geoffmunn Mar 2, 2026
184cacf
Update mul_mat_vec_q_switch_type to include ids_stride parameter in Q…
geoffmunn Mar 2, 2026
1de093b
Fix whitespace issues in HIFI_BUILD_GUIDE.md and quantize.cpp
geoffmunn Mar 2, 2026
1ac3434
Phase 1 of the TURBO plan completed
geoffmunn Mar 4, 2026
4133d36
Phase 2 and 5 complete
geoffmunn Mar 7, 2026
6d66b07
Phase 4 implemented
geoffmunn Mar 7, 2026
f6d04a9
Phase 5 final bits
geoffmunn Mar 7, 2026
6526b9c
TURBO redesign
geoffmunn Mar 8, 2026
a924783
CUDA TURBO build errors fixed
geoffmunn Mar 8, 2026
d2a256e
Add support for additional TURBO quantization types in CUDA backend
geoffmunn Mar 8, 2026
356f7a8
Refine TURBO quantization type mappings in CUDA backend for improved …
geoffmunn Mar 8, 2026
57baca6
CUDA performance improvements
geoffmunn Mar 9, 2026
49b166d
Update stride calculation in CUDA matrix multiplication for TURBO opt…
geoffmunn Mar 9, 2026
51561e4
Fix kernel mapping for Q3_K TURBO implementation in Metal backend and…
geoffmunn Mar 14, 2026
e6656fb
TURBO renamed to LITE
geoffmunn Mar 15, 2026
5c24e4c
Merge pull request #41 from geoffmunn/TURBO
geoffmunn Mar 15, 2026
c1db190
Merge remote-tracking branch 'upstream/master'
geoffmunn Mar 15, 2026
bb131cf
Refactor tensor type handling in llama_tensor_get_type_impl to simpli…
geoffmunn Mar 15, 2026
3e7b4cf
Merge branch 'ggml-org:master' into master
geoffmunn Mar 20, 2026
70965fa
Replace 'bc' with 'awk' for arithmetic operations
geoffmunn Apr 11, 2026
40d639d
Update project configuration and documentation
geoffmunn Apr 11, 2026
246af4e
Refactor GGUF parameter handling in Gemma3Model and Gemma4Model
geoffmunn Apr 11, 2026
f383f85
Revert Gemma3/Gemma4 GGUF parameter refactoring to fix uniform PPL
geoffmunn Apr 12, 2026
0deb61f
Files deleted
geoffmunn Apr 12, 2026
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
15 changes: 15 additions & 0 deletions .gitignore
Original file line number Diff line number Diff line change
Expand Up @@ -145,3 +145,18 @@ poetry.toml
/.windsurf/
# emscripten
a.out.*
wikitext-2-raw/wikitext-2-raw/wiki.test.raw
wikitext-2-raw/wikitext-2-raw/wiki.train.raw
wikitext-2-raw/wikitext-2-raw/wiki.valid.raw
Qwen3-1.7B/.gitattributes
Qwen3-1.7B/config.json
Qwen3-1.7B/generation_config.json
Qwen3-1.7B/LICENSE
Qwen3-1.7B/merges.txt
Qwen3-1.7B/model-00001-of-00002.safetensors
Qwen3-1.7B/model-00002-of-00002.safetensors
Qwen3-1.7B/model.safetensors.index.json
Qwen3-1.7B/README.md
Qwen3-1.7B/tokenizer_config.json
Qwen3-1.7B/tokenizer.json
Qwen3-1.7B/vocab.json
110 changes: 0 additions & 110 deletions AGENTS.md

This file was deleted.

195 changes: 0 additions & 195 deletions CONTRIBUTING.md

This file was deleted.

Loading