Upload folder using huggingface_hub
Browse files- config.json +7 -16
- generation_config.json +1 -1
- recipe.yaml +1 -0
config.json
CHANGED
|
@@ -3,22 +3,17 @@
|
|
| 3 |
"LlamaForCausalLM"
|
| 4 |
],
|
| 5 |
"attention_bias": false,
|
| 6 |
-
"attention_dropout": 0.0,
|
| 7 |
"bos_token_id": 1,
|
| 8 |
-
"dtype": "bfloat16",
|
| 9 |
"eos_token_id": 2,
|
| 10 |
-
"head_dim": 64,
|
| 11 |
"hidden_act": "silu",
|
| 12 |
"hidden_size": 2048,
|
| 13 |
"initializer_range": 0.02,
|
| 14 |
"intermediate_size": 5632,
|
| 15 |
"max_position_embeddings": 2048,
|
| 16 |
-
"mlp_bias": false,
|
| 17 |
"model_type": "llama",
|
| 18 |
"num_attention_heads": 32,
|
| 19 |
"num_hidden_layers": 22,
|
| 20 |
"num_key_value_heads": 4,
|
| 21 |
-
"pad_token_id": null,
|
| 22 |
"pretraining_tp": 1,
|
| 23 |
"quantization_config": {
|
| 24 |
"config_groups": {
|
|
@@ -30,18 +25,14 @@
|
|
| 30 |
"Linear"
|
| 31 |
],
|
| 32 |
"weights": {
|
| 33 |
-
"actorder": null,
|
| 34 |
-
"block_structure": null,
|
| 35 |
"dynamic": false,
|
| 36 |
"group_size": 16,
|
| 37 |
"num_bits": 4,
|
| 38 |
"observer": "memoryless_minmax",
|
| 39 |
-
"observer_kwargs": {},
|
| 40 |
"scale_dtype": "torch.float8_e4m3fn",
|
| 41 |
"strategy": "tensor_group",
|
| 42 |
"symmetric": true,
|
| 43 |
-
"type": "float"
|
| 44 |
-
"zp_dtype": null
|
| 45 |
}
|
| 46 |
}
|
| 47 |
},
|
|
@@ -51,19 +42,19 @@
|
|
| 51 |
"lm_head"
|
| 52 |
],
|
| 53 |
"kv_cache_scheme": null,
|
|
|
|
| 54 |
"quant_method": "compressed-tensors",
|
| 55 |
"quantization_status": "compressed",
|
| 56 |
"sparsity_config": {},
|
| 57 |
"transform_config": {},
|
| 58 |
-
"version": "0.
|
| 59 |
},
|
| 60 |
"rms_norm_eps": 1e-05,
|
| 61 |
-
"
|
| 62 |
-
|
| 63 |
-
"rope_type": "default"
|
| 64 |
-
},
|
| 65 |
"tie_word_embeddings": false,
|
| 66 |
-
"
|
|
|
|
| 67 |
"use_cache": true,
|
| 68 |
"vocab_size": 32000
|
| 69 |
}
|
|
|
|
| 3 |
"LlamaForCausalLM"
|
| 4 |
],
|
| 5 |
"attention_bias": false,
|
|
|
|
| 6 |
"bos_token_id": 1,
|
|
|
|
| 7 |
"eos_token_id": 2,
|
|
|
|
| 8 |
"hidden_act": "silu",
|
| 9 |
"hidden_size": 2048,
|
| 10 |
"initializer_range": 0.02,
|
| 11 |
"intermediate_size": 5632,
|
| 12 |
"max_position_embeddings": 2048,
|
|
|
|
| 13 |
"model_type": "llama",
|
| 14 |
"num_attention_heads": 32,
|
| 15 |
"num_hidden_layers": 22,
|
| 16 |
"num_key_value_heads": 4,
|
|
|
|
| 17 |
"pretraining_tp": 1,
|
| 18 |
"quantization_config": {
|
| 19 |
"config_groups": {
|
|
|
|
| 25 |
"Linear"
|
| 26 |
],
|
| 27 |
"weights": {
|
|
|
|
|
|
|
| 28 |
"dynamic": false,
|
| 29 |
"group_size": 16,
|
| 30 |
"num_bits": 4,
|
| 31 |
"observer": "memoryless_minmax",
|
|
|
|
| 32 |
"scale_dtype": "torch.float8_e4m3fn",
|
| 33 |
"strategy": "tensor_group",
|
| 34 |
"symmetric": true,
|
| 35 |
+
"type": "float"
|
|
|
|
| 36 |
}
|
| 37 |
}
|
| 38 |
},
|
|
|
|
| 42 |
"lm_head"
|
| 43 |
],
|
| 44 |
"kv_cache_scheme": null,
|
| 45 |
+
"moe_sparsity_config": null,
|
| 46 |
"quant_method": "compressed-tensors",
|
| 47 |
"quantization_status": "compressed",
|
| 48 |
"sparsity_config": {},
|
| 49 |
"transform_config": {},
|
| 50 |
+
"version": "0.20.2a20261028"
|
| 51 |
},
|
| 52 |
"rms_norm_eps": 1e-05,
|
| 53 |
+
"rope_scaling": null,
|
| 54 |
+
"rope_theta": 10000.0,
|
|
|
|
|
|
|
| 55 |
"tie_word_embeddings": false,
|
| 56 |
+
"torch_dtype": "bfloat16",
|
| 57 |
+
"transformers_version": "4.35.0",
|
| 58 |
"use_cache": true,
|
| 59 |
"vocab_size": 32000
|
| 60 |
}
|
generation_config.json
CHANGED
|
@@ -3,5 +3,5 @@
|
|
| 3 |
"eos_token_id": 2,
|
| 4 |
"max_length": 2048,
|
| 5 |
"pad_token_id": 0,
|
| 6 |
-
"transformers_version": "5.
|
| 7 |
}
|
|
|
|
| 3 |
"eos_token_id": 2,
|
| 4 |
"max_length": 2048,
|
| 5 |
"pad_token_id": 0,
|
| 6 |
+
"transformers_version": "5.21.0"
|
| 7 |
}
|
recipe.yaml
CHANGED
|
@@ -5,3 +5,4 @@ default_stage:
|
|
| 5 |
ignore: [lm_head, 're:.*mlp.gate[.].*']
|
| 6 |
scheme: NVFP4A16
|
| 7 |
bypass_divisibility_checks: false
|
|
|
|
|
|
| 5 |
ignore: [lm_head, 're:.*mlp.gate[.].*']
|
| 6 |
scheme: NVFP4A16
|
| 7 |
bypass_divisibility_checks: false
|
| 8 |
+
requires_calibration_data: false
|