Banaxi-Tech
/

BananaMind-2.1-Lite-AdamW

Text Generation

bananamind21_lite_25m

partially-looped

Model card Files Files and versions

BananaMind-2.1-Lite-AdamW

440 MB

Ctrl+K

Ctrl+K

1 contributor

History: 24 commits

Banaxi-Tech's picture

Save 100% checkpoint at 75,000,000,000 tokens

8ac0325 verified about 1 month ago

.gitattributes

1.58 kB
Save 85% checkpoint at 63,750,799,360 tokens about 1 month ago
README.md

2.55 kB
Initialize BananaMind 2.1 Lite 25M model card about 1 month ago
checkpoint_metadata.json

3.44 kB
Save 100% checkpoint at 75,000,000,000 tokens about 1 month ago
config.json

1.42 kB
Save 5% checkpoint at 3,750,232,064 tokens about 1 month ago
configuration_bananamind21_lite_25m.py

5.31 kB
Add BananaMind 2.1 Lite 25M training package about 1 month ago
curriculum_75b.py

6 kB
Add BananaMind 2.1 Lite 25M training package about 1 month ago
generation_config.json

127 Bytes
Save 5% checkpoint at 3,750,232,064 tokens about 1 month ago
model.safetensors

113 MB
xet

Save 100% checkpoint at 75,000,000,000 tokens about 1 month ago
modeling_bananamind21_lite_25m.py

18.6 kB
Add BananaMind 2.1 Lite 25M training package about 1 month ago
special_tokens_map.json

107 Bytes
Add BananaMind 2 Nano 8K tokenizer about 1 month ago
tokenizer.json

552 kB
Add BananaMind 2 Nano 8K tokenizer about 1 month ago
tokenizer_config.json

183 Bytes
Add BananaMind 2 Nano 8K tokenizer about 1 month ago
train_lite_25m.py

42.4 kB
Add BananaMind 2.1 Lite 25M training package about 1 month ago
training_metrics.jsonl

13 MB
xet

Save 100% checkpoint at 75,000,000,000 tokens about 1 month ago
training_state.pt

313 MB
xet

Save 100% checkpoint at 75,000,000,000 tokens about 1 month ago