toonist/AnuLM-Coder-400M at main

AnuLM-Coder-400M

Ctrl+K

Ctrl+K

1 contributor

History: 7 commits

toonist's picture

Card: how to load with transformers, and why float32 matters

1892298 verified 18 days ago

.gitattributes

1.52 kB
initial commit 18 days ago
README.md

15.1 kB
Card: how to load with transformers, and why float32 matters 18 days ago
config.json

2.34 kB
Load in float32 by default: bfloat16 weights change the router's expert selection and collapse the output 18 days ago
configuration_anulm.py

3.12 kB
Load with transformers: trust_remote_code modelling code, a tokenizers-format tokenizer.json, and auto_map in config.json (weights untouched) 18 days ago
model.py

47.2 kB
Load with transformers: trust_remote_code modelling code, a tokenizers-format tokenizer.json, and auto_map in config.json (weights untouched) 18 days ago
model.safetensors

797 MB
xet

AnuLM v0.1.0 18 days ago
modeling_anulm.py

12.3 kB
Load in float32 by default: bfloat16 weights change the router's expert selection and collapse the output 18 days ago
tokenizer.code32k.json

411 kB
Tokenizer: rename the format id nanosarvam-bpe -> anulm-bpe (merges unchanged; older readers still accept both) 18 days ago
tokenizer.json

1.32 MB
Load with transformers: trust_remote_code modelling code, a tokenizers-format tokenizer.json, and auto_map in config.json (weights untouched) 18 days ago
tokenizer_config.json

232 Bytes
Load with transformers: trust_remote_code modelling code, a tokenizers-format tokenizer.json, and auto_map in config.json (weights untouched) 18 days ago