Training in progress, step 500

README.md

@@ -20,15 +20,15 @@ should probably proofread and complete it, then remove this comment. -->
 
 This model is a fine-tuned version of [data/ap-gpt-j-6b-sft-qlora-04-08](https://huggingface.co/data/ap-gpt-j-6b-sft-qlora-04-08) on the hugodk-sch/aftonposten_title_prefs dataset.
 It achieves the following results on the evaluation set:
-- Loss: 1.6510
-- Rewards/chosen: 0.0280
-- Rewards/rejected: 0.0141
-- Rewards/accuracies: 0.5129
-- Rewards/margins: 0.0139
-- Logps/rejected: -37.4814
-- Logps/chosen: -33.9646
-- Logits/rejected: -2.2329
-- Logits/chosen: -2.2378
+- Loss: 0.9658
+- Rewards/chosen: -0.0301
+- Rewards/rejected: -0.0652
+- Rewards/accuracies: 0.5278
+- Rewards/margins: 0.0352
+- Logps/rejected: -37.6098
+- Logps/chosen: -34.0775
+- Logits/rejected: -2.2182
+- Logits/chosen: -2.2231
 
 ## Model description
 
@@ -57,33 +57,21 @@ The following hyperparameters were used during training:
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_ratio: 0.1
-- num_epochs: 4
+- num_epochs: 1
 
 ### Training results
 
-| Training Loss | Epoch | Step | Logits/chosen | Logits/rejected | Logps/chosen | Logps/rejected | Validation Loss | Rewards/accuracies | Rewards/chosen | Rewards/margins | Rewards/rejected |
-|:-------------:|:-----:|:----:|:-------------:|:---------------:|:------------:|:--------------:|:---------------:|:------------------:|:--------------:|:---------------:|:----------------:|
-| 1.4583        | 0.26  | 100  | -2.2357       | -2.2308         | -34.0303     | -37.5236       | 1.6162          | 0.5245             | 0.0017         | 0.0045          | -0.0028          |
-| 1.279         | 0.52  | 200  | -2.2359       | -2.2311         | -34.0825     | -37.5923       | 1.6100          | 0.5257             | -0.0192        | 0.0111          | -0.0303          |
-| 1.0043        | 0.78  | 300  | -2.2312       | -2.2263         | -34.0845     | -37.6004       | 1.5962          | 0.5166             | -0.0200        | 0.0135          | -0.0335          |
-| 0.7239        | 1.04  | 400  | 1.6461        | -0.0219         | -0.0311      | 0.5341         | 0.0092          | -37.5945           | -34.0893       | -2.2276         | -2.2324          |
-| 0.6061        | 1.3   | 500  | 1.6487        | -0.0274         | -0.0429      | 0.5395         | 0.0155          | -37.6239           | -34.1030       | -2.2282         | -2.2330          |
-| 0.9255        | 1.56  | 600  | 1.5912        | 0.0108          | -0.0119      | 0.5544         | 0.0228          | -37.5464           | -34.0074       | -2.2273         | -2.2321          |
-| 0.8252        | 1.82  | 700  | 1.6334        | 0.0226          | 0.0045       | 0.5216         | 0.0180          | -37.5053           | -33.9781       | -2.2298         | -2.2346          |
-| 0.2848        | 2.08  | 800  | 1.6033        | 0.0153          | -0.0031      | 0.5249         | 0.0184          | -37.5244           | -33.9964       | -2.2313         | -2.2361          |
-| 0.3671        | 2.34  | 900  | 1.6569        | 0.0283          | 0.0177       | 0.5162         | 0.0106          | -37.4723           | -33.9637       | -2.2309         | -2.2358          |
-| 0.3936        | 2.6   | 1000 | 1.6203        | 0.0348          | 0.0187       | 0.5428         | 0.0161          | -37.4698           | -33.9475       | -2.2325         | -2.2374          |
-| 0.3156        | 2.86  | 1100 | 1.6012        | 0.0302          | 0.0108       | 0.5606         | 0.0194          | -37.4896           | -33.9592       | -2.2326         | -2.2375          |
-| 0.2893        | 3.12  | 1200 | 1.5705        | 0.0346          | 0.0103       | 0.5365         | 0.0243          | -37.4909           | -33.9480       | -2.2335         | -2.2383          |
-| 0.277         | 3.38  | 1300 | 1.6102        | 0.0314          | 0.0121       | 0.5403         | 0.0194          | -37.4865           | -33.9559       | -2.2333         | -2.2382          |
-| 0.139         | 3.64  | 1400 | 1.6181        | 0.0273          | 0.0092       | 0.5307         | 0.0181          | -37.4937           | -33.9663       | -2.2333         | -2.2381          |
-| 0.24          | 3.9   | 1500 | 1.6201        | 0.0327          | 0.0149       | 0.5249         | 0.0178          | -37.4793           | -33.9527       | -2.2332         | -2.2381          |
+| Training Loss | Epoch | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
+|:-------------:|:-----:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
+| 0.8961        | 0.26  | 100  | 0.9767          | 0.0065         | -0.0170          | 0.5365             | 0.0235          | -37.5409       | -34.0252     | -2.2266         | -2.2315       |
+| 0.7699        | 0.52  | 200  | 0.9742          | -0.0141        | -0.0400          | 0.5303             | 0.0259          | -37.5737       | -34.0547     | -2.2234         | -2.2282       |
+| 0.6723        | 0.78  | 300  | 0.9761          | -0.0366        | -0.0616          | 0.5299             | 0.0250          | -37.6047       | -34.0868     | -2.2186         | -2.2234       |
 
 
 ### Framework versions
 
-- PEFT 0.10.0
-- Transformers 4.39.0.dev0
+- PEFT 0.8.2
+- Transformers 4.37.2
 - Pytorch 2.1.2+cu121
-- Datasets 2.14.6
+- Datasets 2.17.0
 - Tokenizers 0.15.1

adapter_config.json

@@ -21,12 +21,12 @@
   "revision": null,
   "target_modules": [
     "down_proj",
+    "up_proj",
     "o_proj",
-    "v_proj",
+    "q_proj",
     "k_proj",
-    "up_proj",
-    "gate_proj",
-    "q_proj"
+    "v_proj",
+    "gate_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:2c47bb1a324b5553592841a0c57e589c161272bd4e084e17244ed50fd058977a
+oid sha256:42067a4afd5eea30b7af1f79d96b539218ed034c29e4342836634cae3abd5bf4
 size 176183216

all_results.json

@@ -1,21 +1,21 @@
 {
-    "epoch": 4.0,
-    "eval_logits/chosen": -2.237769603729248,
-    "eval_logits/rejected": -2.232879638671875,
-    "eval_logps/chosen": -33.96460723876953,
-    "eval_logps/rejected": -37.48141860961914,
-    "eval_loss": 1.6509573459625244,
-    "eval_rewards/accuracies": 0.5128737688064575,
-    "eval_rewards/chosen": 0.02797684073448181,
-    "eval_rewards/margins": 0.013894001953303814,
-    "eval_rewards/rejected": 0.014082840643823147,
-    "eval_runtime": 145.2372,
+    "epoch": 1.0,
+    "eval_logits/chosen": -2.2230520248413086,
+    "eval_logits/rejected": -2.2182328701019287,
+    "eval_logps/chosen": -34.07753372192383,
+    "eval_logps/rejected": -37.60981750488281,
+    "eval_loss": 0.9658033847808838,
+    "eval_rewards/accuracies": 0.5278239250183105,
+    "eval_rewards/chosen": -0.03008819743990898,
+    "eval_rewards/margins": 0.03515118733048439,
+    "eval_rewards/rejected": -0.06523937731981277,
+    "eval_runtime": 145.6671,
     "eval_samples": 343,
-    "eval_samples_per_second": 2.362,
-    "eval_steps_per_second": 0.296,
-    "train_loss": 0.36088944433571457,
-    "train_runtime": 10769.3493,
+    "eval_samples_per_second": 2.355,
+    "eval_steps_per_second": 0.295,
+    "train_loss": 0.8438688600218142,
+    "train_runtime": 3250.9917,
     "train_samples": 3079,
-    "train_samples_per_second": 1.144,
-    "train_steps_per_second": 0.143
+    "train_samples_per_second": 0.947,
+    "train_steps_per_second": 0.118
 }

config.json

@@ -18,8 +18,6 @@
   "n_layer": 28,
   "n_positions": 2048,
   "quantization_config": {
-    "_load_in_4bit": true,
-    "_load_in_8bit": false,
     "bnb_4bit_compute_dtype": "float16",
     "bnb_4bit_quant_type": "nf4",
     "bnb_4bit_use_double_quant": false,
@@ -47,7 +45,7 @@
   },
   "tie_word_embeddings": false,
   "tokenizer_class": "GPT2Tokenizer",
-  "transformers_version": "4.39.0.dev0",
+  "transformers_version": "4.37.2",
   "use_cache": true,
   "vocab_size": 50400
 }

eval_results.json

@@ -1,16 +1,16 @@
 {
-    "epoch": 4.0,
-    "eval_logits/chosen": -2.237769603729248,
-    "eval_logits/rejected": -2.232879638671875,
-    "eval_logps/chosen": -33.96460723876953,
-    "eval_logps/rejected": -37.48141860961914,
-    "eval_loss": 1.6509573459625244,
-    "eval_rewards/accuracies": 0.5128737688064575,
-    "eval_rewards/chosen": 0.02797684073448181,
-    "eval_rewards/margins": 0.013894001953303814,
-    "eval_rewards/rejected": 0.014082840643823147,
-    "eval_runtime": 145.2372,
+    "epoch": 1.0,
+    "eval_logits/chosen": -2.2230520248413086,
+    "eval_logits/rejected": -2.2182328701019287,
+    "eval_logps/chosen": -34.07753372192383,
+    "eval_logps/rejected": -37.60981750488281,
+    "eval_loss": 0.9658033847808838,
+    "eval_rewards/accuracies": 0.5278239250183105,
+    "eval_rewards/chosen": -0.03008819743990898,
+    "eval_rewards/margins": 0.03515118733048439,
+    "eval_rewards/rejected": -0.06523937731981277,
+    "eval_runtime": 145.6671,
     "eval_samples": 343,
-    "eval_samples_per_second": 2.362,
-    "eval_steps_per_second": 0.296
+    "eval_samples_per_second": 2.355,
+    "eval_steps_per_second": 0.295
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 4.0,
-    "train_loss": 0.36088944433571457,
-    "train_runtime": 10769.3493,
+    "epoch": 1.0,
+    "train_loss": 0.8438688600218142,
+    "train_runtime": 3250.9917,
     "train_samples": 3079,
-    "train_samples_per_second": 1.144,
-    "train_steps_per_second": 0.143
+    "train_samples_per_second": 0.947,
+    "train_steps_per_second": 0.118
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:07e1cf58470e526e0cd9888eb2e74b50cb325be26bc3b03661761b2566ce04ee
+oid sha256:b715c426f0a7cb99b86fb1b4c90c55353be43921743323373dda6cb1c70085d8
 size 5176