Update config.json
Browse files- config.json +23 -26
config.json
CHANGED
|
@@ -1,29 +1,26 @@
|
|
| 1 |
{
|
| 2 |
-
"
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 4 |
],
|
| 5 |
-
"
|
| 6 |
-
"
|
| 7 |
-
"eos_token_id": 2,
|
| 8 |
-
"hidden_act": "silu",
|
| 9 |
-
"hidden_size": 4096,
|
| 10 |
-
"initializer_range": 0.02,
|
| 11 |
-
"intermediate_size": 14336,
|
| 12 |
-
"max_position_embeddings": 32768,
|
| 13 |
-
"model_type": "mixtral",
|
| 14 |
-
"num_attention_heads": 32,
|
| 15 |
-
"num_experts_per_tok": 2,
|
| 16 |
-
"num_hidden_layers": 32,
|
| 17 |
-
"num_key_value_heads": 8,
|
| 18 |
-
"num_local_experts": 8,
|
| 19 |
-
"output_router_logits": false,
|
| 20 |
-
"rms_norm_eps": 1e-05,
|
| 21 |
-
"rope_theta": 1000000.0,
|
| 22 |
-
"router_aux_loss_coef": 0.02,
|
| 23 |
-
"sliding_window": null,
|
| 24 |
-
"tie_word_embeddings": false,
|
| 25 |
-
"torch_dtype": "bfloat16",
|
| 26 |
-
"transformers_version": "4.36.0.dev0",
|
| 27 |
-
"use_cache": true,
|
| 28 |
-
"vocab_size": 32000
|
| 29 |
}
|
|
|
|
| 1 |
{
|
| 2 |
+
"alpha_pattern": {},
|
| 3 |
+
"auto_mapping": null,
|
| 4 |
+
"bias": "none",
|
| 5 |
+
"fan_in_fan_out": false,
|
| 6 |
+
"inference_mode": true,
|
| 7 |
+
"init_lora_weights": true,
|
| 8 |
+
"layers_pattern": null,
|
| 9 |
+
"layers_to_transform": null,
|
| 10 |
+
"loftq_config": {},
|
| 11 |
+
"lora_alpha": 64,
|
| 12 |
+
"lora_dropout": 0.05,
|
| 13 |
+
"megatron_config": null,
|
| 14 |
+
"megatron_core": "megatron.core",
|
| 15 |
+
"modules_to_save": null,
|
| 16 |
+
"peft_type": "LORA",
|
| 17 |
+
"r": 32,
|
| 18 |
+
"rank_pattern": {},
|
| 19 |
+
"revision": null,
|
| 20 |
+
"target_modules": [
|
| 21 |
+
"k_proj",
|
| 22 |
+
"w2"
|
| 23 |
],
|
| 24 |
+
"task_type": "CAUSAL_LM",
|
| 25 |
+
"use_rslora": false
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 26 |
}
|