Upload model
Browse files- README.md +0 -24
- adapter_config.json +3 -3
- adapter_model.bin +1 -1
README.md
CHANGED
@@ -1,11 +1,6 @@
|
|
1 |
---
|
2 |
library_name: peft
|
3 |
base_model: tiiuae/falcon-7b
|
4 |
-
license: mit
|
5 |
-
datasets:
|
6 |
-
- ck46/hendrycks_math
|
7 |
-
language:
|
8 |
-
- en
|
9 |
---
|
10 |
|
11 |
# Model Card for Model ID
|
@@ -222,22 +217,3 @@ The following `bitsandbytes` quantization config was used during training:
|
|
222 |
|
223 |
|
224 |
- PEFT 0.6.0.dev0
|
225 |
-
## Training procedure
|
226 |
-
|
227 |
-
|
228 |
-
The following `bitsandbytes` quantization config was used during training:
|
229 |
-
- quant_method: bitsandbytes
|
230 |
-
- load_in_8bit: False
|
231 |
-
- load_in_4bit: True
|
232 |
-
- llm_int8_threshold: 6.0
|
233 |
-
- llm_int8_skip_modules: None
|
234 |
-
- llm_int8_enable_fp32_cpu_offload: False
|
235 |
-
- llm_int8_has_fp16_weight: False
|
236 |
-
- bnb_4bit_quant_type: nf4
|
237 |
-
- bnb_4bit_use_double_quant: False
|
238 |
-
- bnb_4bit_compute_dtype: float16
|
239 |
-
|
240 |
-
### Framework versions
|
241 |
-
|
242 |
-
|
243 |
-
- PEFT 0.6.0.dev0
|
|
|
1 |
---
|
2 |
library_name: peft
|
3 |
base_model: tiiuae/falcon-7b
|
|
|
|
|
|
|
|
|
|
|
4 |
---
|
5 |
|
6 |
# Model Card for Model ID
|
|
|
217 |
|
218 |
|
219 |
- PEFT 0.6.0.dev0
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
adapter_config.json
CHANGED
@@ -4,7 +4,7 @@
|
|
4 |
"base_model_name_or_path": "tiiuae/falcon-7b",
|
5 |
"bias": "none",
|
6 |
"fan_in_fan_out": false,
|
7 |
-
"inference_mode":
|
8 |
"init_lora_weights": true,
|
9 |
"layers_pattern": null,
|
10 |
"layers_to_transform": null,
|
@@ -17,9 +17,9 @@
|
|
17 |
"revision": null,
|
18 |
"target_modules": [
|
19 |
"dense",
|
20 |
-
"
|
21 |
"dense_h_to_4h",
|
22 |
-
"
|
23 |
],
|
24 |
"task_type": "CAUSAL_LM"
|
25 |
}
|
|
|
4 |
"base_model_name_or_path": "tiiuae/falcon-7b",
|
5 |
"bias": "none",
|
6 |
"fan_in_fan_out": false,
|
7 |
+
"inference_mode": true,
|
8 |
"init_lora_weights": true,
|
9 |
"layers_pattern": null,
|
10 |
"layers_to_transform": null,
|
|
|
17 |
"revision": null,
|
18 |
"target_modules": [
|
19 |
"dense",
|
20 |
+
"dense_4h_to_h",
|
21 |
"dense_h_to_4h",
|
22 |
+
"query_key_value"
|
23 |
],
|
24 |
"task_type": "CAUSAL_LM"
|
25 |
}
|
adapter_model.bin
CHANGED
@@ -1,3 +1,3 @@
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
-
oid sha256:
|
3 |
size 522285322
|
|
|
1 |
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:c2010d3520c6d50ef36c12171213faceb04fac9877e3d25e1c1951ff6d98ef99
|
3 |
size 522285322
|