Hugging Face's logo

jc-914
/
prism-quantized

Text Classification
Safetensors
sarvam_mla
decision-model
classification
calibration
int4
gptq
quantized
Mixture of Experts
custom_code
8-bit precision
Model card Files
xet
 Community
prism-quantized
File size: 2,182 Bytes
2d2f883
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
{
  "architectures": [
    "SarvamMLAForCausalLM"
  ],
  "attention_dropout": 0.0,
  "attn_implementation": null,
  "auto_map": {
    "AutoConfig": "configuration_sarvam_moe.SarvamMLAConfig",
    "AutoModel": "modeling_sarvam_moe.SarvamMLAModel",
    "AutoModelForCausalLM": "modeling_sarvam_moe.SarvamMLAForCausalLM"
  },
  "default_theta": 10000.0,
  "dtype": "bfloat16",
  "embedding_dropout": 0.0,
  "eos_token_id": 1,
  "first_k_dense_replace": 1,
  "head_dim": 576,
  "hidden_act": "silu",
  "hidden_size": 4096,
  "initializer_range": 0.006,
  "intermediate_size": 16384,
  "kv_lora_rank": 512,
  "max_position_embeddings": 131072,
  "model_type": "sarvam_mla",
  "moe_intermediate_size": 2048,
  "moe_router_enable_expert_bias": true,
  "num_attention_heads": 64,
  "num_experts": 128,
  "num_experts_per_tok": 8,
  "num_hidden_layers": 32,
  "num_shared_experts": 1,
  "output_dropout": 0.0,
  "output_router_logits": false,
  "pad_token_id": 0,
  "q_head_dim": 192,
  "qk_nope_head_dim": 128,
  "qk_rope_head_dim": 64,
  "rms_norm_eps": 1e-06,
  "rope_scaling": {
    "beta_fast": 32,
    "beta_slow": 1,
    "factor": 40,
    "mscale": 1.0,
    "mscale_all_dim": 1.0,
    "original_max_position_embeddings": 4096,
    "type": "deepseek_yarn"
  },
  "rope_theta": 10000.0,
  "routed_scaling_factor": 2.5,
  "tie_word_embeddings": false,
  "transformers_version": "4.57.2",
  "use_cache": false,
  "use_qk_norm": true,
  "v_head_dim": 128,
  "vocab_size": 262144,
  "torch_dtype": "bfloat16",
  "quantization": {
    "routed_experts": {
      "format": "int4",
      "method": "GPTQ (damp 0.01, MSE clip search)",
      "symmetric": true,
      "group_size": 128,
      "scale_dtype": "float32",
      "packing": "uint8 [E, N, K/2]; in each 128-group byte j = elem j (low nibble) | elem j+64 (high nibble); value = nibble - 8",
      "calibration": "262144 tokens of Prism-format requests over multilingual Wikipedia (11 languages)"
    },
    "other_projections": {
      "format": "float8_e4m3fn",
      "weight_scale": "per output row (absmax / 448)",
      "note": "unchanged from sarvamai/prism"
    },
    "embeddings_norms": "bfloat16",
    "router": "float32"
  }
}