Upload phi-tiny-moe-sft checkpoint (step 32)
Browse files- adapter_config.json +56 -0
- adapter_model.safetensors +3 -0
- chat_template.jinja +4 -0
- tokenizer.json +0 -0
- tokenizer_config.json +16 -0
adapter_config.json
ADDED
|
@@ -0,0 +1,56 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"alora_invocation_tokens": null,
|
| 3 |
+
"alpha_pattern": {
|
| 4 |
+
".*\\.gate_up_proj": 64
|
| 5 |
+
},
|
| 6 |
+
"arrow_config": null,
|
| 7 |
+
"auto_mapping": null,
|
| 8 |
+
"base_model_name_or_path": "microsoft/Phi-tiny-MoE-instruct",
|
| 9 |
+
"bias": "none",
|
| 10 |
+
"corda_config": null,
|
| 11 |
+
"ensure_weight_tying": false,
|
| 12 |
+
"eva_config": null,
|
| 13 |
+
"exclude_modules": null,
|
| 14 |
+
"fan_in_fan_out": false,
|
| 15 |
+
"inference_mode": true,
|
| 16 |
+
"init_lora_weights": true,
|
| 17 |
+
"layer_replication": null,
|
| 18 |
+
"layers_pattern": null,
|
| 19 |
+
"layers_to_transform": null,
|
| 20 |
+
"loftq_config": {},
|
| 21 |
+
"lora_alpha": 32,
|
| 22 |
+
"lora_bias": false,
|
| 23 |
+
"lora_dropout": 0.0,
|
| 24 |
+
"lora_ga_config": null,
|
| 25 |
+
"megatron_config": null,
|
| 26 |
+
"megatron_core": "megatron.core",
|
| 27 |
+
"modules_to_save": null,
|
| 28 |
+
"peft_type": "LORA",
|
| 29 |
+
"peft_version": "0.19.1",
|
| 30 |
+
"qalora_group_size": 16,
|
| 31 |
+
"r": 16,
|
| 32 |
+
"rank_pattern": {
|
| 33 |
+
".*\\.gate_up_proj": 32
|
| 34 |
+
},
|
| 35 |
+
"revision": null,
|
| 36 |
+
"target_modules": [
|
| 37 |
+
"v_proj",
|
| 38 |
+
"q_proj",
|
| 39 |
+
"up_proj",
|
| 40 |
+
"o_proj",
|
| 41 |
+
"k_proj",
|
| 42 |
+
"down_proj",
|
| 43 |
+
"gate_proj"
|
| 44 |
+
],
|
| 45 |
+
"target_parameters": [
|
| 46 |
+
"experts.gate_up_proj",
|
| 47 |
+
"experts.down_proj",
|
| 48 |
+
"router.weight"
|
| 49 |
+
],
|
| 50 |
+
"task_type": "CAUSAL_LM",
|
| 51 |
+
"trainable_token_indices": null,
|
| 52 |
+
"use_bdlora": null,
|
| 53 |
+
"use_dora": false,
|
| 54 |
+
"use_qalora": false,
|
| 55 |
+
"use_rslora": false
|
| 56 |
+
}
|
adapter_model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:acb465beeecc906893dec27846ee403ae4b3bf3edced26b0a62e1e619c9d4c38
|
| 3 |
+
size 528575792
|
chat_template.jinja
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{% for message in messages %}{{'<|' + message['role'] + '|>' + '
|
| 2 |
+
' + message['content'] + '<|end|>
|
| 3 |
+
' }}{% endfor %}{% if add_generation_prompt %}{{ '<|assistant|>
|
| 4 |
+
' }}{% else %}{{ eos_token }}{% endif %}
|
tokenizer.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
tokenizer_config.json
ADDED
|
@@ -0,0 +1,16 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "<s>",
|
| 4 |
+
"clean_up_tokenization_spaces": false,
|
| 5 |
+
"eos_token": "<|endoftext|>",
|
| 6 |
+
"is_local": false,
|
| 7 |
+
"legacy": false,
|
| 8 |
+
"local_files_only": true,
|
| 9 |
+
"model_max_length": 4096,
|
| 10 |
+
"pad_token": "<|endoftext|>",
|
| 11 |
+
"padding_side": "left",
|
| 12 |
+
"sp_model_kwargs": {},
|
| 13 |
+
"tokenizer_class": "TokenizersBackend",
|
| 14 |
+
"unk_token": "<unk>",
|
| 15 |
+
"use_default_system_prompt": false
|
| 16 |
+
}
|