Hugging Face's logo

Spaces:
sandz7
/
loki
Runtime error

App Files Community
loki

Commit History

Renamed project
24f840d
verified

sandz7 commited on

Update README.md
6be716d
verified

sandz7 commited on

steps added
df3f00c

sandz7 commited on

added steps
83d06cb

sandz7 commited on

placed in the delta.content in text and appended that to outputs
39eda37

sandz7 commited on

instanced gpt-4o functions
86917e5

sandz7 commited on

placed the output in list comprehension and just returned the output from output_list
2e99ee0

sandz7 commited on

Removed the spaces in the join so there won't be extra space
6def1b9

sandz7 commited on

placed a space on .join inside output_list
930773d
verified

sandz7 commited on

used a filter instead of .join() twice
8a691c0

sandz7 commited on

joined the output again in output_list
a14c9f3

sandz7 commited on

placed strip(' ') with a space
f123c86

sandz7 commited on

strip() on after the join()
2ed6783

sandz7 commited on

striped and joined the list into a string for output_list
5a708e7

sandz7 commited on

Show the first 15 items in output_list
622f3ea

sandz7 commited on

removed (str) from the output_list function
8899970
verified

sandz7 commited on

placed outputs in the .join() inside the llama_generation
d9a7da0

sandz7 commited on

added streamer in the generate_kwargs and placed params in streamer
aaf89eb

sandz7 commited on

placed a function to convert the list into str from [0]
9078da1

sandz7 commited on

steps
4b9950d

sandz7 commited on

used the params inside .generate()
a4a9909

sandz7 commited on

added login for HF
99033fb

sandz7 commited on

removed thread and text stream as we want the whole output for UI
d5cc98b

sandz7 commited on

removed thread and text streaming as we just need the whole output placed on ui
722a52a

sandz7 commited on

replaced the terminator values to origin and replaced the transformer to instruct version
a0c513d
verified

sandz7 commited on

update
46c2ef7

sandz7 commited on

only added the eos_token_id to terminators
ba43494

sandz7 commited on

update
320e0f4

sandz7 commited on

update
7764f6a

sandz7 commited on

update
b0831b2

sandz7 commited on

removed string on convert_tokens_to_ids
3385018

sandz7 commited on

placed timeout on streamer
405ba08

sandz7 commited on

added prints for logs
b0fb40d

sandz7 commited on

view if eos_token_id is set
3e7a6d0

sandz7 commited on

made a list over the example list
e8bc467

sandz7 commited on

added terminators, params on generations and a thread with steamer to finalize also a sliding feature on UI
76a6e88

sandz7 commited on

reduced top_p to 0.8
10a4f12

sandz7 commited on

changed temperature to 0.5 for more focused response
b96b90a

sandz7 commited on

tuned the parameters for generation
89477ee

sandz7 commited on

reworded header
e4851f5

sandz7 commited on

reworded the header
678f5ce

sandz7 commited on

max new tokens on generation 256
9633993

sandz7 commited on

update on input_ids
516209b

sandz7 commited on

added a newline in header
9d8fdc0

sandz7 commited on

header with user input, and grabbed the first response of output_ids[0]
db79da7

sandz7 commited on

added 2 parameters for llama_generation
92c5d55

sandz7 commited on

added another argument in output_ids
c6e3099

sandz7 commited on

placed input_ids in cuda and just added the vector for input_ids on generate() for model
ec72fe8

sandz7 commited on

placed just the generation into ui for llama3-8b
b28bc26

sandz7 commited on

Added amphisbeana
e83f85a

sandz7 commited on