Given that now the multi chat templating logic itself is used to
apply chat templating/tagging to a single chat message, so give
flexibility of deciding whether global tags if any should be
applied or not wrt the core tagging logic.
examples/main inturn updated to not apply global tags if any wrt
the system message. Also the user messages already dont apply
global tags if any, as its currently implemented to build on the
existing in-prefix/suffix and anitprompt flow.
This should be ok, given that there is a version of the chat tmpl
meta data already included with the library.
So only if user wants to change the chat template info wrt a existing
model/template-standard or add a new one, then there is need to
pass a json file with info for that model/standard.
Have merged master branch has of 20240510IST12XY with chaton_v3
branch.
As part of same had to update the flow in examples/main/main.cpp
wrt conversion related commit in master branch and my chaton related
commits in this branch.
* Revert "Revert "llava : add support for moondream vision language model (#6899)""
This reverts commit 9da243b36a.
* Fix num_positions and embeddings initialization
* convert-hf : begin refactoring write_tensor
* convert : upgrade to sentencepiece v0.2.0
* convert-hf : remove unused n_dims in extra_*_tensors
* convert-hf : simplify MoE weights stacking
* convert-hf : flake8 linter doesn't like semicolons
* convert-hf : allow unusual model part names
For example, loading `model-00001-of-00001.safetensors` now works.
* convert-hf : fix stacking MoE expert tensors
`torch.stack` and `torch.cat` don't do the same thing.
* convert-hf : fix Mamba conversion
Tested to work even with a SentencePiece-based tokenizer.
* convert : use a string for the SentencePiece tokenizer path
* convert-hf : display tensor shape
* convert-hf : convert norms to f32 by default
* convert-hf : sort model part names
`os.listdir` is said to list files in arbitrary order.
Sorting the file names should let "model-00009-of-00042.safetensors"
be loaded before "model-00010-of-00042.safetensors".
* convert-hf : use an ABC for Model again
It seems Protocol can't be used as a statically type-checked ABC,
because its subclasses also can't be instantiated. (why did it seem to work?)
At least there's still a way to throw an error when forgetting to define
the `model_arch` property of any registered Model subclasses.
* convert-hf : use a plain class for Model, and forbid direct instantiation
There are no abstract methods used anyway,
so using ABC isn't really necessary.
* convert-hf : more consistent formatting of cmdline args
* convert-hf : align the message logged for converted tensors
* convert-hf : fix Refact conversion
* convert-hf : save memory with lazy evaluation
* convert-hf : flake8 doesn't like lowercase L as a variable name
* convert-hf : remove einops requirement for InternLM2
* convert-hf : faster model parts loading
Instead of pre-loading them all into a dict, iterate on the tensors
in the model parts progressively as needed in Model.write_tensors
Conversion for some architectures relies on checking for the presence
of specific tensor names, so for multi-part models, the weight map is read
from the relevant json file to quickly get these names up-front.
* convert-hf : minor changes for consistency
* gguf-py : add tqdm as a dependency
It's small, and used for a progress bar
in GGUFWriter.write_tensors_to_file
* Added themes support with two sample themes and a favicon.
* Newline
* Newline
* Newline
* Trailing whitespace
* Increased opacity for contrast
* Increase opacity.
Check actions cancelled for some other priority job and I can't seem to manually re-run them, so MOAR OPACITY
* Opacity action trigger.
Trying to re-trigger the cancelled action.
* One more opacity adjustment
This Actions pipeline is failing for random issues.
* Delete examples/server/themes/buttons_top/completion.js
This will be served from the static string built-in to server.
* Delete examples/server/themes/buttons_top/index.js
This will be served from the static string built-in to server.
* Delete examples/server/themes/wild/completion.js
This will be served from the static string built-in to server.
* Delete examples/server/themes/buttons_top/json-schema-to-grammar.mjs
This will be served from the static string built-in to server.
* Delete examples/server/themes/wild/index.js
This will be served from the static string built-in to server.
* Delete examples/server/themes/wild/json-schema-to-grammar.mjs
This will be served from the static string built-in to server.
* Replaced underscore.
* Introduce bfloat16 support
Many models on Hugging Face (e.g. Mistral, TinyLLaMA) use bfloat16 as
their canonical floating point format.
┌sign
│
│ ┌exponent
│ │
│ │ ┌mantissa
│ │ │
│┌──┴───┐┌─┴───┐
0b0000000000000000 brain16
This encoding has the same number of exponent bits as float32. That
makes conversion relatively straightforward, even in the absence of
hardware support. For example, converting brain16 to binary32 means
simply shifting 16 bits to the left.
┌sign
│
│ ┌exponent
│ │
│ │ ┌mantissa
│ │ │
│┌──┴───┐┌─┴───────────────────┐
0b00000000000000000000000000000000 IEEE binary32
The issue is that converting bf16 to fp16 can result in information
loss. Only 13% of bf16 numbers can be precisely represented in fp16
which in practice ends up being 99.71% of Mistral 7b v0.2's weights
however there is currently no way other than fp32 to get the others
┌sign
│
│ ┌exponent
│ │
│ │ ┌mantissa
│ │ │
│┌─┴─┐┌─┴──────┐
0b0000000000000000 IEEE binary16
This change fixes that, by adding a bf16 data type to GGML. Support
for CPU inference has been implemented along with optimizations for
the AVX2, AVX512, and AVX512BF16 ISAs. Perplexity on Mistral 7b 0.2
improves somewhere around -0.0024 to -0.0046 compared to using fp16
* Remove GGML code that's not needed
* Minimize the GGML API surface area for BF16
* Remove bf16 luts
* Make the GGML header look nicer
* Fix documentation
* Apply ggerganov's fixes for test-backend-ops
* Add BF16 code for new ggml_validate_row_data() function
* Fixed save_imatrix to match old behaviour for MoE
This fix is simple and clear, but unnecessarily doubles the memory overhead..
* Fixed missing idx variable
* Unconditionally increment ncall
Co-authored-by: slaren <slarengh@gmail.com>
* Fixed 2 bugs in save_imatrix()
- Fixed segfault bug because the counts vector needed to be created.
- Fixed pre-existing bug didn't actually add to the counts for "--combine" option.
* ncall needs summing too
* Trailing whitespace
---------
Co-authored-by: slaren <slarengh@gmail.com>
* Update log text (EOS to EOG)
The log text "found EOS" is no longer always correct, here, because there is now an is-EOG check that also returns true for EOT.
* Improve log msg. further by using "an" instead of "some".
As suggested, to avoid misunderstanding (no multiple EOG tokens found, just one).
Have looked at tokenizer_config.json, jinja file and default
hardcoded template in llama.cpp.
This is also one of the models where a Global BoS is needed.
NOTE: Have taken the liberty to also add a SYSTEM: prefix wrt
system message, even thou default vicuna doesnt seem to need, but
vicuna-orca seems to need, so that both models can be driven from
same chat template config. I am assuming the system prefix should
not create any problem even in default vicuna, however if it does
create a problem one can duplicate the existing vicuna block in
chaton_meta.json and make the system prefix empty in it.
The first model seen, based on templates added till now into meta
json file, that needs a Global Begin.
From tokenizer_config json file, it appears like even system role
should have a appropriate prefix, unlike what is seen in hardcoded
default chat apply template of llama.cpp and chat jinja template.
Warn if something not starting with double quote is being treated
as a string.
Show some examples of invalid floating point values wrt this
logics floating point determination code
Add missing begin and end fields wrt deepseek-coder assistant in
chaton_meta.json.
Idea is to avoid json library dependency by adding a simple text
based config file support.
Add a c api wrapper for a single message tagging scenario.
Inturn to match convention followed by existing chat_apply_template
code, make it return the size expected of the tagged message string
buffer. Update internal single logic to help with same.
Explicitly check if tmpl specified is available in the loaded json
or not and then return a error if not found.
Fix a oversight wrt key name.
Add a alert in case if passed meta json file contains begin(BoS)
wrt assistant role, similar to check for end (EoS) wrt user role.
Bcas normally both (ie EoS wrt User and BoS wrt Assistant) shouldnt
be needed.
Update main wrt begin & prefix and suffix & end addition.
Wrt llama2
* add bos wrt llama2 system and user begins, but not assistant
* split system suffix into suffix and end, and add systemuser-system
flags so that end can be avoided wrt system+user message combo
* add eos wrt assistant end
* With these potentially this should work with main and server flows
Wrt llama3
* add empty begin, end fields and systemuser-system flags
* This should potentially work with main and server flows
Was looking at the tokenized vector, and noticed that the EOS
mentioned by existing chat_apply_template of llama.cpp, is different
from what I noticed in tokenizer_config.json of deepseek llm, so
I have added two entries
* "deepseek-alt" which matches llama.cpp's chat_apply_template and
* "deepseek" which matches that in tokenizer_config.json.
This impacts the assistant suffix and reverse prompt entries.
CasOfThis: Need to look into other entries which I added previously
at a later time. However as the default logic should be picking the
EOS from model file, so I assume reverse-prompt being outofsync,
may not matter beyond a limit, potentially.
HanishKVC
Andrei
slaren
Ahmet Zeer
compilade
Johannes Gäßler
Georgi Gerganov
JohnnyB
Dawid Potocki
Johan
Xuan Son Nguyen
Justine Tunney
jukofyork
Kyle Mistele
RhinoDevel
omahs