Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

README.md

@@ -6,7 +6,7 @@ tags:
 - alignment-handbook
 - generated_from_trainer
 datasets:
-- simonycl/Llama-3-8B-Instruct-ultrafeedback-judge-5-annotate
+- simonycl/Meta-Llama-3-8B-Instruct_ultrafeedback-Meta-Llama-3-8B-Instruct-annotate-start-0-end-1.0-judge-5
 model-index:
 - name: llama-3-8b-instruct-agg-judge
   results: []
@@ -17,17 +17,17 @@ should probably proofread and complete it, then remove this comment. -->
 
 # llama-3-8b-instruct-agg-judge
 
-This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the simonycl/Llama-3-8B-Instruct-ultrafeedback-judge-5-annotate dataset.
+This model is a fine-tuned version of [meta-llama/Meta-Llama-3-8B-Instruct](https://huggingface.co/meta-llama/Meta-Llama-3-8B-Instruct) on the simonycl/Meta-Llama-3-8B-Instruct_ultrafeedback-Meta-Llama-3-8B-Instruct-annotate-start-0-end-1.0-judge-5 dataset.
 It achieves the following results on the evaluation set:
-- Loss: 0.5049
-- Rewards/chosen: -1.4456
-- Rewards/rejected: -2.1220
-- Rewards/accuracies: 0.7933
-- Rewards/margins: 0.6764
-- Logps/rejected: -356.5251
-- Logps/chosen: -307.2105
-- Logits/rejected: -1.1659
-- Logits/chosen: -0.9705
+- Loss: 0.6390
+- Rewards/chosen: -1.0532
+- Rewards/rejected: -1.3037
+- Rewards/accuracies: 0.6057
+- Rewards/margins: 0.2506
+- Logps/rejected: -280.7787
+- Logps/chosen: -256.8969
+- Logits/rejected: -1.4905
+- Logits/chosen: -1.5260
 
 ## Model description
 
@@ -48,13 +48,13 @@ More information needed
 The following hyperparameters were used during training:
 - learning_rate: 5e-07
 - train_batch_size: 1
-- eval_batch_size: 1
+- eval_batch_size: 2
 - seed: 42
 - distributed_type: multi-GPU
 - num_devices: 4
-- gradient_accumulation_steps: 32
-- total_train_batch_size: 128
-- total_eval_batch_size: 4
+- gradient_accumulation_steps: 16
+- total_train_batch_size: 64
+- total_eval_batch_size: 8
 - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
 - lr_scheduler_type: cosine
 - lr_scheduler_warmup_ratio: 0.1
@@ -64,12 +64,13 @@ The following hyperparameters were used during training:
 
 | Training Loss | Epoch  | Step | Validation Loss | Rewards/chosen | Rewards/rejected | Rewards/accuracies | Rewards/margins | Logps/rejected | Logps/chosen | Logits/rejected | Logits/chosen |
 |:-------------:|:------:|:----:|:---------------:|:--------------:|:----------------:|:------------------:|:---------------:|:--------------:|:------------:|:---------------:|:-------------:|
-| 0.5505        | 0.8528 | 400  | 0.5049          | -1.4456        | -2.1220          | 0.7933             | 0.6764          | -356.5251      | -307.2105    | -1.1659         | -0.9705       |
+| 0.6265        | 0.4264 | 400  | 0.6455          | -0.7831        | -0.9487          | 0.6504             | 0.1655          | -245.2767      | -229.8961    | -1.3679         | -1.4091       |
+| 0.6053        | 0.8529 | 800  | 0.6390          | -1.0532        | -1.3037          | 0.6057             | 0.2506          | -280.7787      | -256.8969    | -1.4905         | -1.5260       |
 
 
 ### Framework versions
 
-- Transformers 4.44.2
+- Transformers 4.45.1
 - Pytorch 2.4.0+cu121
 - Datasets 2.21.0
-- Tokenizers 0.19.1
+- Tokenizers 0.20.0

all_results.json

@@ -1,22 +1,22 @@
 {
-    "epoch": 0.9999333733093477,
-    "eval_logits/chosen": -0.9731917977333069,
-    "eval_logits/rejected": -1.166278600692749,
-    "eval_logps/chosen": -306.48046875,
-    "eval_logps/rejected": -356.2023010253906,
-    "eval_loss": 0.503829836845398,
-    "eval_rewards/accuracies": 0.7933906316757202,
-    "eval_rewards/chosen": -1.4382753372192383,
-    "eval_rewards/margins": 0.6804661750793457,
-    "eval_rewards/rejected": -2.118741512298584,
-    "eval_runtime": 11557.6821,
-    "eval_samples": 60035,
-    "eval_samples_per_second": 5.194,
-    "eval_steps_per_second": 1.299,
+    "epoch": 1.0,
+    "eval_logits/chosen": -1.532152771949768,
+    "eval_logits/rejected": -1.4972456693649292,
+    "eval_logps/chosen": -261.3055114746094,
+    "eval_logps/rejected": -286.0805358886719,
+    "eval_loss": 0.6392669677734375,
+    "eval_rewards/accuracies": 0.6036585569381714,
+    "eval_rewards/chosen": -1.0972425937652588,
+    "eval_rewards/margins": 0.2594923973083496,
+    "eval_rewards/rejected": -1.356735110282898,
+    "eval_runtime": 169.8112,
+    "eval_samples": 1962,
+    "eval_samples_per_second": 11.554,
+    "eval_steps_per_second": 1.449,
     "total_flos": 0.0,
-    "train_loss": 0.5891387982409138,
-    "train_runtime": 37343.5856,
-    "train_samples": 60035,
-    "train_samples_per_second": 1.608,
-    "train_steps_per_second": 0.013
+    "train_loss": 0.6256769998495513,
+    "train_runtime": 22377.6313,
+    "train_samples": 60029,
+    "train_samples_per_second": 2.683,
+    "train_steps_per_second": 0.042
 }

config.json

@@ -7,6 +7,7 @@
   "attention_dropout": 0.0,
   "bos_token_id": 128000,
   "eos_token_id": 128009,
+  "head_dim": 128,
   "hidden_act": "silu",
   "hidden_size": 4096,
   "initializer_range": 0.02,
@@ -23,7 +24,7 @@
   "rope_theta": 500000.0,
   "tie_word_embeddings": false,
   "torch_dtype": "bfloat16",
-  "transformers_version": "4.44.2",
+  "transformers_version": "4.45.1",
   "use_cache": true,
   "vocab_size": 128256
 }

eval_results.json

@@ -1,16 +1,16 @@
 {
-    "epoch": 0.9999333733093477,
-    "eval_logits/chosen": -0.9731917977333069,
-    "eval_logits/rejected": -1.166278600692749,
-    "eval_logps/chosen": -306.48046875,
-    "eval_logps/rejected": -356.2023010253906,
-    "eval_loss": 0.503829836845398,
-    "eval_rewards/accuracies": 0.7933906316757202,
-    "eval_rewards/chosen": -1.4382753372192383,
-    "eval_rewards/margins": 0.6804661750793457,
-    "eval_rewards/rejected": -2.118741512298584,
-    "eval_runtime": 11557.6821,
-    "eval_samples": 60035,
-    "eval_samples_per_second": 5.194,
-    "eval_steps_per_second": 1.299
+    "epoch": 1.0,
+    "eval_logits/chosen": -1.532152771949768,
+    "eval_logits/rejected": -1.4972456693649292,
+    "eval_logps/chosen": -261.3055114746094,
+    "eval_logps/rejected": -286.0805358886719,
+    "eval_loss": 0.6392669677734375,
+    "eval_rewards/accuracies": 0.6036585569381714,
+    "eval_rewards/chosen": -1.0972425937652588,
+    "eval_rewards/margins": 0.2594923973083496,
+    "eval_rewards/rejected": -1.356735110282898,
+    "eval_runtime": 169.8112,
+    "eval_samples": 1962,
+    "eval_samples_per_second": 11.554,
+    "eval_steps_per_second": 1.449
 }

generation_config.json

@@ -8,5 +8,5 @@
   "max_length": 4096,
   "temperature": 0.6,
   "top_p": 0.9,
-  "transformers_version": "4.44.2"
+  "transformers_version": "4.45.1"
 }

model-00001-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:aadc0d089c01c51929020ecb50c2681dbe21c6798b2542a82bb3fb4a7391f490
+oid sha256:052823a54f52d6374734679dfad3cef0a00d65296dc4dc7681029c79a9ed8d84
 size 4976698672

model-00002-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:799708995ad93f68aa9f73c9fdd0382bf183da7bd6496f320b063ee28ce7f144
+oid sha256:4dfad25c7871229156ac68fb0889049ba24c21802b6fd9d74ada4c66de1d6cf9
 size 4999802720

model-00003-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:948d4df220f4c7139f9b504c8eec9bd66b959994c3cdf878bbff15e5dbbd62dd
+oid sha256:2d235a811d13a087db59f8772e6d6eb03504a2c134a8d4f3e6e9af88a762fec0
 size 4915916176

model-00004-of-00004.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:74cc414c0293f182445de8a80f8c3ddd96095a8d563049d72584d4a715e6831d
+oid sha256:e20b133a4b2022fc82164370fd72f3a001aafc8f651a5f486fe0d132554c0f6b
 size 1168138808

train_results.json

@@ -1,9 +1,9 @@
 {
-    "epoch": 0.9999333733093477,
+    "epoch": 1.0,
     "total_flos": 0.0,
-    "train_loss": 0.5891387982409138,
-    "train_runtime": 37343.5856,
-    "train_samples": 60035,
-    "train_samples_per_second": 1.608,
-    "train_steps_per_second": 0.013
+    "train_loss": 0.6256769998495513,
+    "train_runtime": 22377.6313,
+    "train_samples": 60029,
+    "train_samples_per_second": 2.683,
+    "train_steps_per_second": 0.042
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:9a6681a55e194ed88e0b0738936a1587e12b39b5d17f641aed90732330e377db
-size 7544
+oid sha256:6a76c348c35ea1879fa0299481efe3fff378d2076fd9fceb2667d890b6baea0d
+size 7608