wipe old checkpoint
Browse files- checkpoint/config.json +0 -44
checkpoint/config.json
DELETED
|
@@ -1,44 +0,0 @@
|
|
| 1 |
-
{
|
| 2 |
-
"architectures": [
|
| 3 |
-
"LlamaForCausalLM"
|
| 4 |
-
],
|
| 5 |
-
"attention_bias": false,
|
| 6 |
-
"attention_dropout": 0.0,
|
| 7 |
-
"bos_token_id": 1,
|
| 8 |
-
"capacity_factor_eval": 2.0,
|
| 9 |
-
"capacity_factor_train": 1.25,
|
| 10 |
-
"drop_tks": true,
|
| 11 |
-
"dropout_rate": 0.0,
|
| 12 |
-
"dtype": "float32",
|
| 13 |
-
"eos_token_id": 2,
|
| 14 |
-
"expert_parallel": null,
|
| 15 |
-
"gated": true,
|
| 16 |
-
"head_dim": 64,
|
| 17 |
-
"hidden_act": "silu",
|
| 18 |
-
"hidden_size": 768,
|
| 19 |
-
"initializer_range": 0.02,
|
| 20 |
-
"intermediate_size": 2048,
|
| 21 |
-
"layer_norm_epsilon": 1e-06,
|
| 22 |
-
"max_position_embeddings": 2048,
|
| 23 |
-
"min_capacity": 4,
|
| 24 |
-
"mlp_bias": false,
|
| 25 |
-
"model_type": "llama",
|
| 26 |
-
"moe_layer_interval": 4,
|
| 27 |
-
"noisy_policy": null,
|
| 28 |
-
"num_attention_heads": 12,
|
| 29 |
-
"num_experts": 16,
|
| 30 |
-
"num_hidden_layers": 12,
|
| 31 |
-
"num_key_value_heads": 12,
|
| 32 |
-
"pad_token_id": 0,
|
| 33 |
-
"pretraining_tp": 1,
|
| 34 |
-
"rms_norm_eps": 1e-06,
|
| 35 |
-
"rope_parameters": {
|
| 36 |
-
"rope_theta": 10000.0,
|
| 37 |
-
"rope_type": "default"
|
| 38 |
-
},
|
| 39 |
-
"tie_word_embeddings": false,
|
| 40 |
-
"topk": 2,
|
| 41 |
-
"transformers_version": "5.2.0",
|
| 42 |
-
"use_cache": true,
|
| 43 |
-
"vocab_size": 256384
|
| 44 |
-
}
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|