Training in progress, step 40

axolotl_config.yaml

@@ -1,5 +1,5 @@
-base_model: unsloth/OpenHermes-2.5-Mistral-7B
-batch_size: 16
+base_model: facebook/opt-125m
+batch_size: 128
 bf16: true
 chat_template: tokenizer_default_fallback_alpaca
 datasets:
@@ -25,7 +25,7 @@ learning_rate: 0.0002
 logging_steps: 10
 lr_scheduler: cosine
 max_steps: 2500
-micro_batch_size: 2
+micro_batch_size: 16
 model_type: AutoModelForCausalLM
 num_epochs: 100
 optimizer: adamw_bnb_8bit
@@ -36,13 +36,13 @@ sample_packing: true
 save_steps: 40
 save_total_limit: 2
 sequence_len: 2048
-tokenizer_type: LlamaTokenizerFast
+tokenizer_type: GPT2TokenizerFast
 torch_dtype: bf16
 trust_remote_code: true
 val_set_size: 0.1
 wandb_entity: ''
 wandb_mode: online
-wandb_name: unsloth/OpenHermes-2.5-Mistral-7B-argilla/databricks-dolly-15k-curated-en
+wandb_name: facebook/opt-125m-argilla/databricks-dolly-15k-curated-en
 wandb_project: Gradients-On-Demand
 wandb_run: your_name
 wandb_runid: default

config.json

@@ -1,28 +1,31 @@
 {
-  "_name_or_path": "unsloth/OpenHermes-2.5-Mistral-7B",
+  "_name_or_path": "facebook/opt-125m",
+  "_remove_final_layer_norm": false,
+  "activation_dropout": 0.0,
+  "activation_function": "relu",
   "architectures": [
-    "MistralForCausalLM"
+    "OPTForCausalLM"
   ],
   "attention_dropout": 0.0,
-  "bos_token_id": 1,
-  "eos_token_id": 32000,
-  "head_dim": 128,
-  "hidden_act": "silu",
-  "hidden_size": 4096,
-  "initializer_range": 0.02,
-  "intermediate_size": 14336,
-  "max_position_embeddings": 32768,
-  "model_type": "mistral",
-  "num_attention_heads": 32,
-  "num_hidden_layers": 32,
-  "num_key_value_heads": 8,
-  "pad_token_id": 0,
-  "rms_norm_eps": 1e-05,
-  "rope_theta": 10000.0,
-  "sliding_window": 4096,
-  "tie_word_embeddings": false,
+  "bos_token_id": 2,
+  "do_layer_norm_before": true,
+  "dropout": 0.1,
+  "enable_bias": true,
+  "eos_token_id": 2,
+  "ffn_dim": 3072,
+  "hidden_size": 768,
+  "init_std": 0.02,
+  "layer_norm_elementwise_affine": true,
+  "layerdrop": 0.0,
+  "max_position_embeddings": 2048,
+  "model_type": "opt",
+  "num_attention_heads": 12,
+  "num_hidden_layers": 12,
+  "pad_token_id": 1,
+  "prefix": "</s>",
   "torch_dtype": "bfloat16",
   "transformers_version": "4.48.1",
   "use_cache": false,
-  "vocab_size": 32002
+  "vocab_size": 50265,
+  "word_embed_proj_dim": 768
 }

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:87e0feac30a2b7a0f3658bfbd058914e27b5492b737a7f09f8b8bcbcb91e35df
-size 136062744
+oid sha256:020fb30959dcfbcac82451f41a438ae05e3b42759d7aef54dbd8b1b18fa92a0b
+size 250490408

special_tokens_map.json

@@ -1,29 +1,29 @@
 {
   "bos_token": {
-    "content": "<s>",
+    "content": "</s>",
     "lstrip": false,
-    "normalized": false,
+    "normalized": true,
     "rstrip": false,
     "single_word": false
   },
   "eos_token": {
-    "content": "<|im_end|>",
+    "content": "</s>",
     "lstrip": false,
-    "normalized": false,
+    "normalized": true,
     "rstrip": false,
     "single_word": false
   },
   "pad_token": {
-    "content": "<unk>",
+    "content": "<pad>",
     "lstrip": false,
-    "normalized": false,
+    "normalized": true,
     "rstrip": false,
     "single_word": false
   },
   "unk_token": {
-    "content": "<unk>",
+    "content": "</s>",
     "lstrip": false,
-    "normalized": false,
+    "normalized": true,
     "rstrip": false,
     "single_word": false
   }

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:04222cd76979c181cd3f72c3bf6982fe2a09d9f4b8f23d82902efde18f1d0668
-size 3506125
+oid sha256:4a80800503667fe0bd639ad10c33879f747ba1582f369b06abb21f3f65d5ad3b
+size 3558658

tokenizer_config.json

@@ -1,20 +1,11 @@
 {
   "add_bos_token": true,
-  "add_eos_token": false,
-  "add_prefix_space": true,
+  "add_prefix_space": false,
   "added_tokens_decoder": {
-    "0": {
-      "content": "<unk>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
     "1": {
-      "content": "<s>",
+      "content": "<pad>",
       "lstrip": false,
-      "normalized": false,
+      "normalized": true,
       "rstrip": false,
       "single_word": false,
       "special": true
@@ -22,43 +13,21 @@
     "2": {
       "content": "</s>",
       "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "32000": {
-      "content": "<|im_end|>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
-    },
-    "32001": {
-      "content": "<|im_start|>",
-      "lstrip": false,
-      "normalized": false,
+      "normalized": true,
       "rstrip": false,
       "single_word": false,
       "special": true
     }
   },
-  "additional_special_tokens": [],
-  "bos_token": "<s>",
-  "chat_template": "{% for message in messages %}{{'<|im_start|>' + message['role'] + '\n' + message['content'] + '<|im_end|>' + '\n'}}{% endfor %}{% if add_generation_prompt %}{{ '<|im_start|>assistant\n' }}{% endif %}",
+  "bos_token": "</s>",
+  "chat_template": "{% for message in messages %}{% if message['role'] == 'user' %}{{ '### Instruction: ' + message['content'] + '\n\n' }}{% elif message['role'] == 'assistant' %}{{ '### Response: ' + message['content'] + eos_token}}{% endif %}{% endfor %}",
   "clean_up_tokenization_spaces": false,
-  "eos_token": "<|im_end|>",
+  "eos_token": "</s>",
+  "errors": "replace",
   "extra_special_tokens": {},
-  "legacy": true,
-  "model_max_length": 32768,
-  "pad_token": "<unk>",
-  "padding_side": "right",
-  "sp_model_kwargs": {},
-  "spaces_between_special_tokens": false,
-  "tokenizer_class": "LlamaTokenizer",
-  "trust_remote_code": false,
-  "unk_token": "<unk>",
-  "use_default_system_prompt": true,
+  "model_max_length": 1000000000000000019884624838656,
+  "pad_token": "<pad>",
+  "tokenizer_class": "GPT2Tokenizer",
+  "unk_token": "</s>",
   "use_fast": true
 }

training_args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:53ebcd35271d188c087a5f00ee35959f144cc8fbcdfc0d1744678c3f065510bd
-size 6776
+oid sha256:f3d5f1e240ecc8a87b6c5953fd6de84410c7aff5f9646db311ea17ff72210b93
+size 6840