Hugging Face's logo

AlessaC
/
qwen2p5b_em_badmed_r32

Transformers
Safetensors
unsloth
Model card Files
xet
 Community
AlessaC commited on
Commit
94b4c2a
·
verified ·
1 Parent(s): 91c8fbb

Training in progress, step 200

Browse files
Files changed (9) hide show
  1. adapter_config.json +5 -5
  2. adapter_model.safetensors +1 -1
  3. kernel_metrics/K_0_metadata.json +12 -0
  4. kernel_metrics/K_0_reference_ntk.npz +3 -0
  5. kernel_metrics/kernel_behavior_analysis.txt +2 -2
  6. kernel_metrics/kernel_metrics.json +52 -0
  7. kernel_metrics/kernel_summary.json +8 -0
  8. tokenizer.json +2 -2
  9. training_args.bin +1 -1
adapter_config.json CHANGED
@@ -24,13 +24,13 @@
24
  "rank_pattern": {},
25
  "revision": null,
26
  "target_modules": [
27
- "up_proj",
28
- "q_proj",
29
- "o_proj",
30
  "gate_proj",
31
- "v_proj",
 
 
 
32
  "k_proj",
33
- "down_proj"
34
  ],
35
  "task_type": "CAUSAL_LM",
36
  "trainable_token_indices": null,
 
24
  "rank_pattern": {},
25
  "revision": null,
26
  "target_modules": [
 
 
 
27
  "gate_proj",
28
+ "down_proj",
29
+ "o_proj",
30
+ "q_proj",
31
+ "up_proj",
32
  "k_proj",
33
+ "v_proj"
34
  ],
35
  "task_type": "CAUSAL_LM",
36
  "trainable_token_indices": null,
adapter_model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:d9b4aa40a85d432dad3679601170806960f62af71ae85ea8a9e180eae2579605
3
  size 70430032
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c61392ceee0f17408472923a95258c9efc36e6ec7a27191e3d6143f7ab1117b3
3
  size 70430032
kernel_metrics/K_0_metadata.json ADDED
@@ -0,0 +1,12 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "description": "Reference Neural Tangent Kernel K_0 computed at initialization \u03b8_0",
3
+ "kernel_mode": "signgd",
4
+ "scalar_mode": "logprob",
5
+ "shape": [
6
+ 20,
7
+ 20
8
+ ],
9
+ "trace": 20.0,
10
+ "frobenius_norm": 20.0,
11
+ "num_eval_prompts": 20
12
+ }
kernel_metrics/K_0_reference_ntk.npz ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9d1a3f18e6cf922630fee0c443a71e08b4e1cae53af6390dbf2b4cf136466455
3
+ size 612
kernel_metrics/kernel_behavior_analysis.txt CHANGED
@@ -3,8 +3,8 @@ KERNEL BEHAVIOR ANALYSIS
3
  Based on Malladi et al. (2023) Section 6.2
4
  ================================================================================
5
 
6
- Training steps analyzed: 200 to 600
7
- Number of checkpoints: 3
8
  Number of eval prompts: 5
9
 
10
  --------------------------------------------------------------------------------
 
3
  Based on Malladi et al. (2023) Section 6.2
4
  ================================================================================
5
 
6
+ Training steps analyzed: 200 to 794
7
+ Number of checkpoints: 4
8
  Number of eval prompts: 5
9
 
10
  --------------------------------------------------------------------------------
kernel_metrics/kernel_metrics.json CHANGED
@@ -154,5 +154,57 @@
154
  ],
155
  "linearization_error": NaN,
156
  "kernel_distance": 0.0
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
157
  }
158
  ]
 
154
  ],
155
  "linearization_error": NaN,
156
  "kernel_distance": 0.0
157
+ },
158
+ {
159
+ "step": 794,
160
+ "kernel_matrix": [
161
+ [
162
+ 0.0,
163
+ 0.0,
164
+ 0.0,
165
+ 0.0,
166
+ 0.0
167
+ ],
168
+ [
169
+ 0.0,
170
+ 0.0,
171
+ 0.0,
172
+ 0.0,
173
+ 0.0
174
+ ],
175
+ [
176
+ 0.0,
177
+ 0.0,
178
+ 0.0,
179
+ 0.0,
180
+ 0.0
181
+ ],
182
+ [
183
+ 0.0,
184
+ 0.0,
185
+ 0.0,
186
+ 0.0,
187
+ 0.0
188
+ ],
189
+ [
190
+ 0.0,
191
+ 0.0,
192
+ 0.0,
193
+ 0.0,
194
+ 0.0
195
+ ]
196
+ ],
197
+ "kernel_trace": 0.0,
198
+ "kernel_frobenius_norm": 0.0,
199
+ "mean_gradient_norm": 0.0,
200
+ "gradient_norms": [
201
+ 0.0,
202
+ 0.0,
203
+ 0.0,
204
+ 0.0,
205
+ 0.0
206
+ ],
207
+ "linearization_error": NaN,
208
+ "kernel_distance": 0.0
209
  }
210
  ]
kernel_metrics/kernel_summary.json CHANGED
@@ -22,5 +22,13 @@
22
  "mean_gradient_norm": 0.0,
23
  "linearization_error": NaN,
24
  "kernel_distance": 0.0
 
 
 
 
 
 
 
 
25
  }
26
  ]
 
22
  "mean_gradient_norm": 0.0,
23
  "linearization_error": NaN,
24
  "kernel_distance": 0.0
25
+ },
26
+ {
27
+ "step": 794,
28
+ "kernel_trace": 0.0,
29
+ "kernel_frobenius_norm": 0.0,
30
+ "mean_gradient_norm": 0.0,
31
+ "linearization_error": NaN,
32
+ "kernel_distance": 0.0
33
  }
34
  ]
tokenizer.json CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:913950e4971737031da511cdd1b410daae4566f62eb845b3975bca5a102323d8
3
- size 11421995
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
3
+ size 11421896
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:e22247c25709b0f4e40d1ffb31e892da76f69e4b8d03e9c958c5683242674554
3
  size 6033
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:fe7450d7563a8cb3169fe4ae712391177fbcdebc4e5777360e085f1a96a38850
3
  size 6033