Commit
•
f7b808d
1
Parent(s):
deac39a
Upload 4 files
Browse filesBreaking upload into chunks due to failures.
- config.json +37 -0
- mergekit_moe_config.yml +13 -0
- model-00001-of-00023.safetensors +3 -0
- model.safetensors.index.json +0 -0
config.json
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "01-ai/Yi-Coder-9B-Chat",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"MixtralForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": false,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 1,
|
| 9 |
+
"eos_token_id": 7,
|
| 10 |
+
"head_dim": 128,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 4096,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 11008,
|
| 15 |
+
"max_position_embeddings": 131072,
|
| 16 |
+
"mlp_bias": false,
|
| 17 |
+
"model_type": "mixtral",
|
| 18 |
+
"num_attention_heads": 32,
|
| 19 |
+
"num_experts_per_tok": 2,
|
| 20 |
+
"num_hidden_layers": 48,
|
| 21 |
+
"num_key_value_heads": 4,
|
| 22 |
+
"num_local_experts": 8,
|
| 23 |
+
"output_router_logits": false,
|
| 24 |
+
"pad_token_id": 0,
|
| 25 |
+
"pretraining_tp": 1,
|
| 26 |
+
"rms_norm_eps": 1e-05,
|
| 27 |
+
"rope_scaling": null,
|
| 28 |
+
"rope_theta": 10000000,
|
| 29 |
+
"router_aux_loss_coef": 0.001,
|
| 30 |
+
"router_jitter_noise": 0.0,
|
| 31 |
+
"sliding_window": null,
|
| 32 |
+
"tie_word_embeddings": false,
|
| 33 |
+
"torch_dtype": "bfloat16",
|
| 34 |
+
"transformers_version": "4.44.2",
|
| 35 |
+
"use_cache": false,
|
| 36 |
+
"vocab_size": 64000
|
| 37 |
+
}
|
mergekit_moe_config.yml
ADDED
|
@@ -0,0 +1,13 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
base_model: 01-ai/Yi-Coder-9B-Chat
|
| 2 |
+
gate_mode: random
|
| 3 |
+
dtype: bfloat16
|
| 4 |
+
experts:
|
| 5 |
+
- source_model: 01-ai/Yi-Coder-9B-Chat
|
| 6 |
+
- source_model: 01-ai/Yi-Coder-9B-Chat
|
| 7 |
+
- source_model: 01-ai/Yi-Coder-9B-Chat
|
| 8 |
+
- source_model: 01-ai/Yi-Coder-9B-Chat
|
| 9 |
+
- source_model: 01-ai/Yi-Coder-9B-Chat
|
| 10 |
+
- source_model: 01-ai/Yi-Coder-9B-Chat
|
| 11 |
+
- source_model: 01-ai/Yi-Coder-9B-Chat
|
| 12 |
+
- source_model: 01-ai/Yi-Coder-9B-Chat
|
| 13 |
+
# and then train the sucker!
|
model-00001-of-00023.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f295b50889d6a51096b6fe0872f0d688cb4d1ab31d1489bc5e14a8a587221ca2
|
| 3 |
+
size 4913667584
|
model.safetensors.index.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|