JohnP1
/

d1a-e2b-mlx-q8

Text Classification

8-bit precision

Model card Files Files and versions

4.76 GB

Ctrl+K

Ctrl+K

1 contributor

History: 5 commits

JohnP1's picture

v0.2: add the base's vision and audio encoders (media/) for photos and voice

841ba19 verified 7 days ago

media
v0.2: add the base's vision and audio encoders (media/) for photos and voice 7 days ago
.gitattributes

1.57 kB
D1A-E2B v0.1 MLX q8 (per-layer embeddings 4-bit), parity-checked 9 days ago
README.md

2.74 kB
v0.2: add the base's vision and audio encoders (media/) for photos and voice 7 days ago
config.json

4.96 kB
D1A-E2B v0.1 MLX q8 (per-layer embeddings 4-bit), parity-checked 9 days ago
d1a_config.json

782 Bytes
Calibration temperature 1.52 (fitted on the study's calibration split; the published head.pt had been left at 1.0) 8 days ago
head.safetensors

3.15 MB
xet

D1A-E2B v0.2 (two epochs) MLX q8, parity-checked 9 days ago
model.safetensors

3.74 GB
xet

D1A-E2B v0.2 (two epochs) MLX q8, parity-checked 9 days ago
model.safetensors.index.json

95.6 kB
D1A-E2B v0.1 MLX q8 (per-layer embeddings 4-bit), parity-checked 9 days ago
tokenizer.json

32.2 MB
xet

D1A-E2B v0.1 MLX q8 (per-layer embeddings 4-bit), parity-checked 9 days ago
tokenizer_config.json

1.53 kB
D1A-E2B v0.1 MLX q8 (per-layer embeddings 4-bit), parity-checked 9 days ago