PEFT
Safetensors
llama
axolotl
Generated from Trainer
minpeter commited on
Commit
251067c
·
verified ·
1 Parent(s): 6a83eef

Upload folder using huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +14 -6
README.md CHANGED
@@ -130,7 +130,15 @@ It achieves the following results on the evaluation set:
130
 
131
  ## Model description
132
 
133
- More information needed
 
 
 
 
 
 
 
 
134
 
135
  ## Intended uses & limitations
136
 
@@ -158,11 +166,11 @@ The following hyperparameters were used during training:
158
 
159
  ### Training results
160
 
161
- | Training Loss | Epoch | Step | Validation Loss |
162
- |:-------------:|:------:|:----:|:---------------:|
163
- | 0.8199 | 0.0072 | 1 | 0.6489 |
164
- | 0.0898 | 0.3381 | 47 | 0.2228 |
165
- | 0.2114 | 0.6763 | 94 | 0.2159 |
166
 
167
 
168
  ### Framework versions
 
130
 
131
  ## Model description
132
 
133
+
134
+ | Test Category | Adapter Accuracy | Base Model Accuracy | Improvement |
135
+ | ----------------- | ---------------- | ------------------- | ----------- |
136
+ | irrelevance | 76.25% | 72.08% | +4.17% |
137
+ | parallel_multiple | 89.50% | 10.00% | +79.50% |
138
+ | parallel | 89.50% | 11.50% | +78.00% |
139
+ | simple | 92.75% | 24.75% | +68.00% |
140
+ | multiple | 93.50% | 20.00% | +73.50% |
141
+
142
 
143
  ## Intended uses & limitations
144
 
 
166
 
167
  ### Training results
168
 
169
+ | Training Loss | Epoch | Step | Validation Loss |
170
+ | :-----------: | :----: | :---: | :-------------: |
171
+ | 0.8199 | 0.0072 | 1 | 0.6489 |
172
+ | 0.0898 | 0.3381 | 47 | 0.2228 |
173
+ | 0.2114 | 0.6763 | 94 | 0.2159 |
174
 
175
 
176
  ### Framework versions