Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Tags: chraac/llama.cpp

Tags

b9631

Toggle b9631's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
cli : fix not copying preserved tokens (ggml-org#24258)

b9630

Toggle b9630's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
Add cohere2moe to llama-vocab for TINY_AYA (ggml-org#24601)

b9628

Toggle b9628's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
add sycl to check-release (ggml-org#24583)

b9627

Toggle b9627's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
ui : fix llama-ui-embed crash when no asset dir is given (ggml-org#24597

)

b9626

Toggle b9626's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
Add arch support for cohere2-MoE (ggml-org#24260)

* Add arch support for cohere2-MoE

* Removed redundant gating_func checks

* Changed ffn lookup to prefer prefix_dense_intermediate_size

* Renamed arch to cohere2moe

* Removed redundant lmhead check and chat template changes

* Removed lm_head.weight check from modify tensors, load output tensor not required, fallback to token_embd.weight

* Changed to (routed+shared)*0.5 for shared expert combined avg

* fixed sliding_window_pattern issue and pattern

* Fixed transformers crash 'first_k_dense_replace' error

* Remove comment

* Removed cohere2-moe as a tokenizer type and kept as tiny_aya.  Renamed North-Mini-Code-1.0.

* Fixed MTP fail, changed to use iSWA

* Fixed remaining todos: cohere2moe renamed, changed swa parsing to use get_key_or_arr, removed extra get_arr use

* Force metadata usage

Co-authored-by: Sigbjørn Skjæret <[email protected]>

* Remove Cohere2 checkpoint comment

Co-authored-by: Sigbjørn Skjæret <[email protected]>

* Remove MTP comment

Co-authored-by: Sigbjørn Skjæret <[email protected]>

* Regenerate cohere2moe tokenizer hash

* Add cohere2moe to Llama Model Saver supported list

* Check for zerobios tensors and add support for Command to use LayerNorm

* Map expert_selection_fn to sigmoid in base.py instead of command.py

* use bools for foundnorm/foundnormrms

Co-authored-by: Sigbjørn Skjæret <[email protected]>

---------

Co-authored-by: Sigbjørn Skjæret <[email protected]>

b9625

Toggle b9625's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
jinja : fix negative step slice with start/stop values (ggml-org#24580)

b9624

Toggle b9624's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
ui: build-time gzip compression (ggml-org#24571)

* ui: keep original file name and path

* fix nocache

* ui: build-time gzip compression

b9623

Toggle b9623's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
jinja : fix split and replace with empty first arg (ggml-org#24574)

* fix split and replace with empty first arg

* fix reserve size

b9622

Toggle b9622's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
vulkan: support non-contig unary/glu ops (ggml-org#24215)

* vulkan: support non-contig unary/glu ops

Change unary/glu ops to pass in all strides and use fastdiv for the index
calculation. Put all unary ops in one file, similar to glu, to share the
code. codex went ahead and added expm1 without me asking, but I had to
make it do a real precision analysis rather than just making stuff up.

unary.comp initially couldn't use generic_unary_head because there wasn't
space for xielu's additional constants. Fixing this required packing the
fastdiv 'L' values.

* attempt to workaround compiler bug

* resolve conflict from ggml-org#23991

* use expm1

b9621

Toggle b9621's commit message

Verified

This commit was created on GitHub.com and signed with GitHub’s verified signature.
ui: keep original file name and path (ggml-org#24568)

* ui: keep original file name and path

* fix nocache