issues
search
ReaLLMASIC
/
nanoGPT
The simplest, fastest repository for training/finetuning medium-sized GPTs.
MIT License
24
stars
18
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
Updated scripts to extract zh snac tokens
#312
xinyixuu
opened
1 day ago
0
update CUDA source for ReLU forward pass
#311
Hrancheng
opened
6 days ago
0
Test
#310
gkielian
opened
1 week ago
0
Add scripts compatible with smoltalk dataset
#309
klei22
opened
1 week ago
0
Reimplement Tokenization Script
#308
gkielian
closed
1 week ago
1
Create PR for DXLS to PPA (or add to PR)
#307
gkielian
opened
1 week ago
0
Add additional models for Validation Loss and Benchmarking
#306
gkielian
opened
1 week ago
0
Create a new directory for XLS
#305
Mars-Cat2023
opened
2 weeks ago
0
Added support for Qwen2 models
#304
mmoffatt2
opened
2 weeks ago
0
Got rid of prev_run_ckpt from config name
#303
mmoffatt2
opened
2 weeks ago
2
Add off-by-one variation for Softmax, as well as temperature and input saturation
#302
klei22
closed
1 week ago
1
Expand chess features
#301
klei22
closed
1 week ago
0
Revise Korean to IPA script to be more robust to inputs
#300
gkielian
closed
1 week ago
1
Add japanese to ipa/romaji
#299
Zhao-Yuting
closed
3 weeks ago
0
cancel
#298
Zhao-Yuting
closed
3 weeks ago
0
Refactor non ipa language targeted words to use [[[[[word]]]]]
#297
gkielian
closed
1 week ago
1
Add attention variant FastAttention from Performer
#296
gkielian
opened
3 weeks ago
1
Add glotcc ko subset
#295
klei22
closed
3 weeks ago
0
Refactor to add activation config
#294
gkielian
closed
3 weeks ago
0
Refactor activations so that we can have config parameters and different activations per layer
#293
gkielian
opened
3 weeks ago
0
Add ICCAD 2024 ConSmax Publication
#292
gkielian
closed
3 weeks ago
0
option for inference with static scale/zero point
#291
mmoffatt2
opened
4 weeks ago
0
RMSNorm Recompute
#290
mmoffatt2
opened
1 month ago
0
Probably useful scripts to profile gpu with gpustats
#289
Hrancheng
opened
1 month ago
1
Added tiny-stories dataset
#288
mmoffatt2
closed
1 month ago
0
Added Lambda "Quantization Level" Parameter
#287
mmoffatt2
closed
3 weeks ago
1
Added ternary quantization option
#286
mmoffatt2
closed
4 weeks ago
1
Print Sample of Activations and Weights during Evaluation
#285
mmoffatt2
opened
1 month ago
0
Add major checkpoints
#284
gkielian
closed
3 weeks ago
1
Added README file and Scripts to Save and Visualizing Quantized Values
#283
mmoffatt2
closed
3 weeks ago
1
Added more symmetric quantization options (per tensor and per channel)
#282
mmoffatt2
opened
1 month ago
0
Add hw efficient and learned GELU variations
#281
gkielian
closed
3 weeks ago
1
Removed Star from gpt2 Argument
#280
mmoffatt2
closed
1 month ago
0
Remove torch from requirements_cpu
#279
gkielian
closed
1 month ago
0
CI Fixes
#278
gkielian
closed
1 month ago
0
Add util for mixed korean and english to ipa
#277
gkielian
closed
1 month ago
0
Add ConSmax Evalution Configuration File and Augmentations
#276
gkielian
closed
1 month ago
0
Added tril for softmax input quantization
#275
mmoffatt2
closed
1 month ago
0
Add support for txt360
#274
klei22
closed
3 weeks ago
0
Removed star from gpt2 argument
#273
mmoffatt2
closed
1 month ago
1
Add support for Attn Variant 'Performer'
#272
gkielian
closed
1 week ago
1
Add flare_finetuning_configuration.json
#271
gkielian
closed
2 months ago
0
Adding script and related files to run tts demo
#270
xinyixuu
opened
2 months ago
1
FiReLU output zeros statistics
#269
mmoffatt2
opened
2 months ago
0
Add FIReLU
#268
gkielian
opened
2 months ago
0
Add gptconf defaults and manual ability to turn off flash attention
#267
gkielian
closed
1 month ago
1
Add csv processing scripts
#266
gkielian
closed
1 month ago
0
Sample.py eval_only Timing
#265
mmoffatt2
closed
4 weeks ago
0
Add learned steering vectors
#264
gkielian
closed
1 month ago
1
Add pos emb finetuning option
#263
gkielian
closed
2 months ago
0
Next