Martial Terran
MartialTerran
AI & ML interests
I, Martial Terran am leading a Group to build solar-powered TimeCapsuleTeacher(TM} GPT-powered laptop computers, to provide Language, Math and Science Education to Non-English-Speaking people of the future in a Post-Apophis World.
Recent Activity
new activity 2 days ago
WeiboAI/VibeThinker-3B:Really looking forward for 9B or 12B variants updated a model 6 days ago
MartialTerran/Method_for_Dynamically_Reducing_Logit_Computation_in_LLMs updated a Space 19 days ago
TimeCapsuleTeacherLLM/READMEOrganizations
Really looking forward for 9B or 12B variants
4
#20 opened 4 days ago
by 
perelmanych
Image Generator modifying Images using AutoEncoder
1
#6 opened 9 months ago
by MartialTerran
404 - page not found
#2 opened 9 months ago
by MartialTerran
Where is SmolLM2_model.py???
2
#1 opened over 1 year ago
by MartialTerran
Size Mismatch in safetensors file
6
#3 opened over 1 year ago
by MartialTerran
DragonAI-Python-SmolLM2_model.py???
3
#1 opened over 1 year ago
by MartialTerran
Under-100M Parameter for detecting 20 Marathi numbers?
3
#1 opened over 1 year ago
by MartialTerran
Error. Crash. "The attention mask is not set and cannot be inferred from input
1
#8 opened over 1 year ago
by MartialTerran
Qwen2 sample model.py does not work.
7
#7 opened over 1 year ago
by MartialTerran
B/c Size Mismatch, Cant use from transformers import LlamaForCausalLM as workaround.
1
#5 opened over 1 year ago
by MartialTerran
Also cant use from transformers import LlamaForCausalLM as a workaround, because of size mismatch.
2
#4 opened over 1 year ago
by MartialTerran
Safetensors size mismatch.
5
#4 opened over 1 year ago
by MartialTerran
Sample Model Script for bfloat16 downloads safetensors parameters files then declares mismatch in their dimensions.
1
#3 opened over 1 year ago
by MartialTerran
Need Help to build a SmolLM2_360M_model.py
1
#2 opened over 1 year ago
by MartialTerran
Distinguishing between speech and non speech
3
#74 opened over 3 years ago
by 
CarelessWhisperer
Phoneme recognition
6
#86 opened about 3 years ago
by 
dg96
Whisper Finetuning - Validation loss is increasing but WER is Decreasing
2
#107 opened over 2 years ago
by 
anahar
Storing Spelling information in LLMs
6
#2 opened over 1 year ago
by MartialTerran
Duplicates in Train set
1
#12 opened over 2 years ago
by 
Qilex
Cleaned data
❤️ 5
3
#15 opened over 2 years ago
by 
ad8e