Hugging Face's logo

jedisct1
/
Mellum2-12B-A2.5B-Instruct-mlx-8bit

Text Generation
MLX
Safetensors
English
mellum
Mixture of Experts
code
conversational
8-bit precision
Model card Files
xet
 Community
Mellum2-12B-A2.5B-Instruct-mlx-8bit / config.json
jedisct1's picture
jedisct1
Add 8-bit MLX quantization
9dde51c verified
Raw History Blame Contribute Delete
8.54 kB
{
"architectures": [
"MellumForCausalLM"
],
"attention_bias": false,
"attention_dropout": 0.0,
"bos_token_id": 0,
"dtype": "bfloat16",
"eos_token_id": [
0,
28
],
"head_dim": 128,
"hidden_act": "silu",
"hidden_size": 2304,
"initializer_range": 0.02,
"intermediate_size": 7168,
"layer_types": [
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention",
"sliding_attention",
"sliding_attention",
"sliding_attention",
"full_attention"
],
"max_position_embeddings": 131072,
"max_window_layers": 0,
"mlp_layer_types": [
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse",
"sparse"
],
"model_type": "mellum",
"moe_intermediate_size": 896,
"norm_topk_prob": true,
"num_attention_heads": 32,
"num_experts": 64,
"num_experts_per_tok": 8,
"num_hidden_layers": 28,
"num_key_value_heads": 4,
"output_router_logits": false,
"quantization": {
"group_size": 64,
"bits": 8,
"mode": "affine",
"model.layers.0.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.1.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.2.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.3.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.4.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.5.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.6.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.7.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.8.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.9.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.10.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.11.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.12.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.13.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.14.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.15.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.16.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.17.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.18.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.19.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.20.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.21.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.22.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.23.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.24.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.25.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.26.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.27.mlp.gate": {
"group_size": 64,
"bits": 8
}
},
"quantization_config": {
"group_size": 64,
"bits": 8,
"mode": "affine",
"model.layers.0.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.1.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.2.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.3.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.4.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.5.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.6.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.7.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.8.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.9.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.10.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.11.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.12.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.13.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.14.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.15.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.16.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.17.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.18.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.19.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.20.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.21.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.22.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.23.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.24.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.25.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.26.mlp.gate": {
"group_size": 64,
"bits": 8
},
"model.layers.27.mlp.gate": {
"group_size": 64,
"bits": 8
}
},
"rms_norm_eps": 1e-06,
"rope_parameters": {
"full_attention": {
"rope_type": "yarn",
"rope_theta": 500000.0,
"factor": 16.0,
"original_max_position_embeddings": 8192,
"beta_fast": 32.0,
"beta_slow": 1.0,
"attention_factor": 1.2772588722239782
},
"sliding_attention": {
"rope_type": "default",
"rope_theta": 500000.0
}
},
"router_aux_loss_coef": 0.001,
"sliding_window": 1024,
"tie_word_embeddings": false,
"transformers_version": "5.8.1",
"use_cache": true,
"vocab_size": 98304
}