End of training
Browse files- config.json +1 -13
config.json
CHANGED
@@ -1,5 +1,5 @@
|
|
1 |
{
|
2 |
-
"_name_or_path": "
|
3 |
"architectures": [
|
4 |
"MistralForCausalLM"
|
5 |
],
|
@@ -15,18 +15,6 @@
|
|
15 |
"num_hidden_layers": 32,
|
16 |
"num_key_value_heads": 8,
|
17 |
"pad_token_id": 2,
|
18 |
-
"quantization_config": {
|
19 |
-
"bnb_4bit_compute_dtype": "float16",
|
20 |
-
"bnb_4bit_quant_type": "nf4",
|
21 |
-
"bnb_4bit_use_double_quant": false,
|
22 |
-
"llm_int8_enable_fp32_cpu_offload": false,
|
23 |
-
"llm_int8_has_fp16_weight": false,
|
24 |
-
"llm_int8_skip_modules": null,
|
25 |
-
"llm_int8_threshold": 6.0,
|
26 |
-
"load_in_4bit": true,
|
27 |
-
"load_in_8bit": false,
|
28 |
-
"quant_method": "bitsandbytes"
|
29 |
-
},
|
30 |
"rms_norm_eps": 1e-05,
|
31 |
"rope_theta": 10000.0,
|
32 |
"sliding_window": 4096,
|
|
|
1 |
{
|
2 |
+
"_name_or_path": "hllj/zephyr-7b-beta-vi-math",
|
3 |
"architectures": [
|
4 |
"MistralForCausalLM"
|
5 |
],
|
|
|
15 |
"num_hidden_layers": 32,
|
16 |
"num_key_value_heads": 8,
|
17 |
"pad_token_id": 2,
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
18 |
"rms_norm_eps": 1e-05,
|
19 |
"rope_theta": 10000.0,
|
20 |
"sliding_window": 4096,
|