Add files using upload-large-folder tool
Browse filesThis view is limited to 50 files because it contains too many changes. See raw diff
- D21/checkpoint-16000/trainer_state.json +0 -0
- D21/checkpoint-6000/tokenizer_config.json +17 -0
- D41/checkpoint-1000/chat_template.jinja +1 -0
- D41/checkpoint-1000/config.json +36 -0
- D41/checkpoint-1000/generation_config.json +12 -0
- D41/checkpoint-1000/tokenizer_config.json +17 -0
- D41/checkpoint-1000/trainer_state.json +0 -0
- D41/checkpoint-11000/chat_template.jinja +1 -0
- D41/checkpoint-11000/tokenizer_config.json +17 -0
- D41/checkpoint-12000/chat_template.jinja +1 -0
- D41/checkpoint-12000/config.json +36 -0
- D41/checkpoint-12000/generation_config.json +12 -0
- D41/checkpoint-12000/tokenizer_config.json +17 -0
- D41/checkpoint-12000/trainer_state.json +0 -0
- D41/checkpoint-2000/chat_template.jinja +1 -0
- D41/checkpoint-2000/config.json +36 -0
- D41/checkpoint-2000/generation_config.json +12 -0
- D41/checkpoint-2000/tokenizer_config.json +17 -0
- D41/checkpoint-2000/trainer_state.json +0 -0
- D41/checkpoint-3000/chat_template.jinja +1 -0
- D41/checkpoint-3000/config.json +36 -0
- D41/checkpoint-3000/generation_config.json +12 -0
- D41/checkpoint-3000/tokenizer_config.json +17 -0
- D41/checkpoint-3000/trainer_state.json +0 -0
- D41/checkpoint-4000/chat_template.jinja +1 -0
- D41/checkpoint-4000/config.json +36 -0
- D41/checkpoint-4000/generation_config.json +12 -0
- D41/checkpoint-4000/tokenizer_config.json +17 -0
- D41/checkpoint-4000/trainer_state.json +0 -0
- D41/checkpoint-5000/chat_template.jinja +1 -0
- D41/checkpoint-5000/config.json +36 -0
- D41/checkpoint-5000/generation_config.json +12 -0
- D41/checkpoint-5000/tokenizer_config.json +17 -0
- D41/checkpoint-5000/trainer_state.json +0 -0
- D41/trainer_log.jsonl +0 -0
- D41/viz_html/layer_01_mean_fixed_canvas_v3/assets/checkpoint-1000/fixed_canvas_v3_w420_h315_dpi140_layer_01_group_00_panel_00_mean.png +0 -0
- D41/viz_html/layer_01_mean_fixed_canvas_v3/assets/checkpoint-1000/fixed_canvas_v3_w420_h315_dpi140_layer_01_group_00_panel_01_mean.png +0 -0
- D41/viz_html/layer_01_mean_fixed_canvas_v3/assets/checkpoint-1000/fixed_canvas_v3_w420_h315_dpi140_layer_01_group_00_panel_02_mean.png +0 -0
- D41/viz_html/layer_01_mean_fixed_canvas_v3/assets/checkpoint-1000/fixed_canvas_v3_w420_h315_dpi140_layer_01_group_01_panel_00_mean.png +0 -0
- D41/viz_html/layer_01_mean_fixed_canvas_v3/assets/checkpoint-1000/fixed_canvas_v3_w420_h315_dpi140_layer_01_group_01_panel_01_mean.png +0 -0
- D41/viz_html/layer_01_mean_fixed_canvas_v3/assets/checkpoint-4000/fixed_canvas_v3_w420_h315_dpi140_layer_01_group_00_panel_00_mean.png +0 -0
- D41/viz_html/layer_01_mean_fixed_canvas_v3/assets/checkpoint-4000/fixed_canvas_v3_w420_h315_dpi140_layer_01_group_01_panel_01_mean.png +0 -0
- D41/viz_html/layer_01_mean_fixed_canvas_v3/assets/checkpoint-5000/fixed_canvas_v3_w420_h315_dpi140_layer_01_group_00_panel_01_mean.png +0 -0
- D63/checkpoint-1000/chat_template.jinja +1 -0
- D63/checkpoint-1000/config.json +36 -0
- D63/checkpoint-1000/generation_config.json +12 -0
- D63/checkpoint-1000/tokenizer_config.json +17 -0
- D63/checkpoint-1000/trainer_state.json +0 -0
- D63/checkpoint-10000/trainer_state.json +0 -0
- D63/checkpoint-11000/trainer_state.json +0 -0
D21/checkpoint-16000/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
D21/checkpoint-6000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "<|begin_of_text|>",
|
| 4 |
+
"clean_up_tokenization_spaces": true,
|
| 5 |
+
"eos_token": "<|end_of_text|>",
|
| 6 |
+
"is_local": true,
|
| 7 |
+
"local_files_only": false,
|
| 8 |
+
"model_input_names": [
|
| 9 |
+
"input_ids",
|
| 10 |
+
"attention_mask"
|
| 11 |
+
],
|
| 12 |
+
"model_max_length": 131072,
|
| 13 |
+
"pad_token": "<|end_of_text|>",
|
| 14 |
+
"padding_side": "right",
|
| 15 |
+
"split_special_tokens": false,
|
| 16 |
+
"tokenizer_class": "TokenizersBackend"
|
| 17 |
+
}
|
D41/checkpoint-1000/chat_template.jinja
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content }}{% endif %}{% endfor %}
|
D41/checkpoint-1000/config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 128000,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"eos_token_id": 128001,
|
| 10 |
+
"head_dim": 64,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2048,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 8192,
|
| 15 |
+
"max_position_embeddings": 131072,
|
| 16 |
+
"mlp_bias": false,
|
| 17 |
+
"model_type": "llama",
|
| 18 |
+
"num_attention_heads": 32,
|
| 19 |
+
"num_hidden_layers": 16,
|
| 20 |
+
"num_key_value_heads": 8,
|
| 21 |
+
"pad_token_id": 128001,
|
| 22 |
+
"pretraining_tp": 1,
|
| 23 |
+
"rms_norm_eps": 1e-05,
|
| 24 |
+
"rope_parameters": {
|
| 25 |
+
"factor": 32.0,
|
| 26 |
+
"high_freq_factor": 4.0,
|
| 27 |
+
"low_freq_factor": 1.0,
|
| 28 |
+
"original_max_position_embeddings": 8192,
|
| 29 |
+
"rope_theta": 500000.0,
|
| 30 |
+
"rope_type": "llama3"
|
| 31 |
+
},
|
| 32 |
+
"tie_word_embeddings": true,
|
| 33 |
+
"transformers_version": "5.6.0",
|
| 34 |
+
"use_cache": false,
|
| 35 |
+
"vocab_size": 128256
|
| 36 |
+
}
|
D41/checkpoint-1000/generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 128000,
|
| 4 |
+
"do_sample": true,
|
| 5 |
+
"eos_token_id": [
|
| 6 |
+
128001
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 128001,
|
| 9 |
+
"temperature": 0.6,
|
| 10 |
+
"top_p": 0.9,
|
| 11 |
+
"transformers_version": "5.6.0"
|
| 12 |
+
}
|
D41/checkpoint-1000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "<|begin_of_text|>",
|
| 4 |
+
"clean_up_tokenization_spaces": true,
|
| 5 |
+
"eos_token": "<|end_of_text|>",
|
| 6 |
+
"is_local": true,
|
| 7 |
+
"local_files_only": false,
|
| 8 |
+
"model_input_names": [
|
| 9 |
+
"input_ids",
|
| 10 |
+
"attention_mask"
|
| 11 |
+
],
|
| 12 |
+
"model_max_length": 131072,
|
| 13 |
+
"pad_token": "<|end_of_text|>",
|
| 14 |
+
"padding_side": "right",
|
| 15 |
+
"split_special_tokens": false,
|
| 16 |
+
"tokenizer_class": "TokenizersBackend"
|
| 17 |
+
}
|
D41/checkpoint-1000/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
D41/checkpoint-11000/chat_template.jinja
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content }}{% endif %}{% endfor %}
|
D41/checkpoint-11000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "<|begin_of_text|>",
|
| 4 |
+
"clean_up_tokenization_spaces": true,
|
| 5 |
+
"eos_token": "<|end_of_text|>",
|
| 6 |
+
"is_local": true,
|
| 7 |
+
"local_files_only": false,
|
| 8 |
+
"model_input_names": [
|
| 9 |
+
"input_ids",
|
| 10 |
+
"attention_mask"
|
| 11 |
+
],
|
| 12 |
+
"model_max_length": 131072,
|
| 13 |
+
"pad_token": "<|end_of_text|>",
|
| 14 |
+
"padding_side": "right",
|
| 15 |
+
"split_special_tokens": false,
|
| 16 |
+
"tokenizer_class": "TokenizersBackend"
|
| 17 |
+
}
|
D41/checkpoint-12000/chat_template.jinja
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content }}{% endif %}{% endfor %}
|
D41/checkpoint-12000/config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 128000,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"eos_token_id": 128001,
|
| 10 |
+
"head_dim": 64,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2048,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 8192,
|
| 15 |
+
"max_position_embeddings": 131072,
|
| 16 |
+
"mlp_bias": false,
|
| 17 |
+
"model_type": "llama",
|
| 18 |
+
"num_attention_heads": 32,
|
| 19 |
+
"num_hidden_layers": 16,
|
| 20 |
+
"num_key_value_heads": 8,
|
| 21 |
+
"pad_token_id": 128001,
|
| 22 |
+
"pretraining_tp": 1,
|
| 23 |
+
"rms_norm_eps": 1e-05,
|
| 24 |
+
"rope_parameters": {
|
| 25 |
+
"factor": 32.0,
|
| 26 |
+
"high_freq_factor": 4.0,
|
| 27 |
+
"low_freq_factor": 1.0,
|
| 28 |
+
"original_max_position_embeddings": 8192,
|
| 29 |
+
"rope_theta": 500000.0,
|
| 30 |
+
"rope_type": "llama3"
|
| 31 |
+
},
|
| 32 |
+
"tie_word_embeddings": true,
|
| 33 |
+
"transformers_version": "5.6.0",
|
| 34 |
+
"use_cache": false,
|
| 35 |
+
"vocab_size": 128256
|
| 36 |
+
}
|
D41/checkpoint-12000/generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 128000,
|
| 4 |
+
"do_sample": true,
|
| 5 |
+
"eos_token_id": [
|
| 6 |
+
128001
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 128001,
|
| 9 |
+
"temperature": 0.6,
|
| 10 |
+
"top_p": 0.9,
|
| 11 |
+
"transformers_version": "5.6.0"
|
| 12 |
+
}
|
D41/checkpoint-12000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "<|begin_of_text|>",
|
| 4 |
+
"clean_up_tokenization_spaces": true,
|
| 5 |
+
"eos_token": "<|end_of_text|>",
|
| 6 |
+
"is_local": true,
|
| 7 |
+
"local_files_only": false,
|
| 8 |
+
"model_input_names": [
|
| 9 |
+
"input_ids",
|
| 10 |
+
"attention_mask"
|
| 11 |
+
],
|
| 12 |
+
"model_max_length": 131072,
|
| 13 |
+
"pad_token": "<|end_of_text|>",
|
| 14 |
+
"padding_side": "right",
|
| 15 |
+
"split_special_tokens": false,
|
| 16 |
+
"tokenizer_class": "TokenizersBackend"
|
| 17 |
+
}
|
D41/checkpoint-12000/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
D41/checkpoint-2000/chat_template.jinja
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content }}{% endif %}{% endfor %}
|
D41/checkpoint-2000/config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 128000,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"eos_token_id": 128001,
|
| 10 |
+
"head_dim": 64,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2048,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 8192,
|
| 15 |
+
"max_position_embeddings": 131072,
|
| 16 |
+
"mlp_bias": false,
|
| 17 |
+
"model_type": "llama",
|
| 18 |
+
"num_attention_heads": 32,
|
| 19 |
+
"num_hidden_layers": 16,
|
| 20 |
+
"num_key_value_heads": 8,
|
| 21 |
+
"pad_token_id": 128001,
|
| 22 |
+
"pretraining_tp": 1,
|
| 23 |
+
"rms_norm_eps": 1e-05,
|
| 24 |
+
"rope_parameters": {
|
| 25 |
+
"factor": 32.0,
|
| 26 |
+
"high_freq_factor": 4.0,
|
| 27 |
+
"low_freq_factor": 1.0,
|
| 28 |
+
"original_max_position_embeddings": 8192,
|
| 29 |
+
"rope_theta": 500000.0,
|
| 30 |
+
"rope_type": "llama3"
|
| 31 |
+
},
|
| 32 |
+
"tie_word_embeddings": true,
|
| 33 |
+
"transformers_version": "5.6.0",
|
| 34 |
+
"use_cache": false,
|
| 35 |
+
"vocab_size": 128256
|
| 36 |
+
}
|
D41/checkpoint-2000/generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 128000,
|
| 4 |
+
"do_sample": true,
|
| 5 |
+
"eos_token_id": [
|
| 6 |
+
128001
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 128001,
|
| 9 |
+
"temperature": 0.6,
|
| 10 |
+
"top_p": 0.9,
|
| 11 |
+
"transformers_version": "5.6.0"
|
| 12 |
+
}
|
D41/checkpoint-2000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "<|begin_of_text|>",
|
| 4 |
+
"clean_up_tokenization_spaces": true,
|
| 5 |
+
"eos_token": "<|end_of_text|>",
|
| 6 |
+
"is_local": true,
|
| 7 |
+
"local_files_only": false,
|
| 8 |
+
"model_input_names": [
|
| 9 |
+
"input_ids",
|
| 10 |
+
"attention_mask"
|
| 11 |
+
],
|
| 12 |
+
"model_max_length": 131072,
|
| 13 |
+
"pad_token": "<|end_of_text|>",
|
| 14 |
+
"padding_side": "right",
|
| 15 |
+
"split_special_tokens": false,
|
| 16 |
+
"tokenizer_class": "TokenizersBackend"
|
| 17 |
+
}
|
D41/checkpoint-2000/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
D41/checkpoint-3000/chat_template.jinja
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content }}{% endif %}{% endfor %}
|
D41/checkpoint-3000/config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 128000,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"eos_token_id": 128001,
|
| 10 |
+
"head_dim": 64,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2048,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 8192,
|
| 15 |
+
"max_position_embeddings": 131072,
|
| 16 |
+
"mlp_bias": false,
|
| 17 |
+
"model_type": "llama",
|
| 18 |
+
"num_attention_heads": 32,
|
| 19 |
+
"num_hidden_layers": 16,
|
| 20 |
+
"num_key_value_heads": 8,
|
| 21 |
+
"pad_token_id": 128001,
|
| 22 |
+
"pretraining_tp": 1,
|
| 23 |
+
"rms_norm_eps": 1e-05,
|
| 24 |
+
"rope_parameters": {
|
| 25 |
+
"factor": 32.0,
|
| 26 |
+
"high_freq_factor": 4.0,
|
| 27 |
+
"low_freq_factor": 1.0,
|
| 28 |
+
"original_max_position_embeddings": 8192,
|
| 29 |
+
"rope_theta": 500000.0,
|
| 30 |
+
"rope_type": "llama3"
|
| 31 |
+
},
|
| 32 |
+
"tie_word_embeddings": true,
|
| 33 |
+
"transformers_version": "5.6.0",
|
| 34 |
+
"use_cache": false,
|
| 35 |
+
"vocab_size": 128256
|
| 36 |
+
}
|
D41/checkpoint-3000/generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 128000,
|
| 4 |
+
"do_sample": true,
|
| 5 |
+
"eos_token_id": [
|
| 6 |
+
128001
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 128001,
|
| 9 |
+
"temperature": 0.6,
|
| 10 |
+
"top_p": 0.9,
|
| 11 |
+
"transformers_version": "5.6.0"
|
| 12 |
+
}
|
D41/checkpoint-3000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "<|begin_of_text|>",
|
| 4 |
+
"clean_up_tokenization_spaces": true,
|
| 5 |
+
"eos_token": "<|end_of_text|>",
|
| 6 |
+
"is_local": true,
|
| 7 |
+
"local_files_only": false,
|
| 8 |
+
"model_input_names": [
|
| 9 |
+
"input_ids",
|
| 10 |
+
"attention_mask"
|
| 11 |
+
],
|
| 12 |
+
"model_max_length": 131072,
|
| 13 |
+
"pad_token": "<|end_of_text|>",
|
| 14 |
+
"padding_side": "right",
|
| 15 |
+
"split_special_tokens": false,
|
| 16 |
+
"tokenizer_class": "TokenizersBackend"
|
| 17 |
+
}
|
D41/checkpoint-3000/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
D41/checkpoint-4000/chat_template.jinja
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content }}{% endif %}{% endfor %}
|
D41/checkpoint-4000/config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 128000,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"eos_token_id": 128001,
|
| 10 |
+
"head_dim": 64,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2048,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 8192,
|
| 15 |
+
"max_position_embeddings": 131072,
|
| 16 |
+
"mlp_bias": false,
|
| 17 |
+
"model_type": "llama",
|
| 18 |
+
"num_attention_heads": 32,
|
| 19 |
+
"num_hidden_layers": 16,
|
| 20 |
+
"num_key_value_heads": 8,
|
| 21 |
+
"pad_token_id": 128001,
|
| 22 |
+
"pretraining_tp": 1,
|
| 23 |
+
"rms_norm_eps": 1e-05,
|
| 24 |
+
"rope_parameters": {
|
| 25 |
+
"factor": 32.0,
|
| 26 |
+
"high_freq_factor": 4.0,
|
| 27 |
+
"low_freq_factor": 1.0,
|
| 28 |
+
"original_max_position_embeddings": 8192,
|
| 29 |
+
"rope_theta": 500000.0,
|
| 30 |
+
"rope_type": "llama3"
|
| 31 |
+
},
|
| 32 |
+
"tie_word_embeddings": true,
|
| 33 |
+
"transformers_version": "5.6.0",
|
| 34 |
+
"use_cache": false,
|
| 35 |
+
"vocab_size": 128256
|
| 36 |
+
}
|
D41/checkpoint-4000/generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 128000,
|
| 4 |
+
"do_sample": true,
|
| 5 |
+
"eos_token_id": [
|
| 6 |
+
128001
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 128001,
|
| 9 |
+
"temperature": 0.6,
|
| 10 |
+
"top_p": 0.9,
|
| 11 |
+
"transformers_version": "5.6.0"
|
| 12 |
+
}
|
D41/checkpoint-4000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "<|begin_of_text|>",
|
| 4 |
+
"clean_up_tokenization_spaces": true,
|
| 5 |
+
"eos_token": "<|end_of_text|>",
|
| 6 |
+
"is_local": true,
|
| 7 |
+
"local_files_only": false,
|
| 8 |
+
"model_input_names": [
|
| 9 |
+
"input_ids",
|
| 10 |
+
"attention_mask"
|
| 11 |
+
],
|
| 12 |
+
"model_max_length": 131072,
|
| 13 |
+
"pad_token": "<|end_of_text|>",
|
| 14 |
+
"padding_side": "right",
|
| 15 |
+
"split_special_tokens": false,
|
| 16 |
+
"tokenizer_class": "TokenizersBackend"
|
| 17 |
+
}
|
D41/checkpoint-4000/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
D41/checkpoint-5000/chat_template.jinja
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content }}{% endif %}{% endfor %}
|
D41/checkpoint-5000/config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 128000,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"eos_token_id": 128001,
|
| 10 |
+
"head_dim": 64,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2048,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 8192,
|
| 15 |
+
"max_position_embeddings": 131072,
|
| 16 |
+
"mlp_bias": false,
|
| 17 |
+
"model_type": "llama",
|
| 18 |
+
"num_attention_heads": 32,
|
| 19 |
+
"num_hidden_layers": 16,
|
| 20 |
+
"num_key_value_heads": 8,
|
| 21 |
+
"pad_token_id": 128001,
|
| 22 |
+
"pretraining_tp": 1,
|
| 23 |
+
"rms_norm_eps": 1e-05,
|
| 24 |
+
"rope_parameters": {
|
| 25 |
+
"factor": 32.0,
|
| 26 |
+
"high_freq_factor": 4.0,
|
| 27 |
+
"low_freq_factor": 1.0,
|
| 28 |
+
"original_max_position_embeddings": 8192,
|
| 29 |
+
"rope_theta": 500000.0,
|
| 30 |
+
"rope_type": "llama3"
|
| 31 |
+
},
|
| 32 |
+
"tie_word_embeddings": true,
|
| 33 |
+
"transformers_version": "5.6.0",
|
| 34 |
+
"use_cache": false,
|
| 35 |
+
"vocab_size": 128256
|
| 36 |
+
}
|
D41/checkpoint-5000/generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 128000,
|
| 4 |
+
"do_sample": true,
|
| 5 |
+
"eos_token_id": [
|
| 6 |
+
128001
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 128001,
|
| 9 |
+
"temperature": 0.6,
|
| 10 |
+
"top_p": 0.9,
|
| 11 |
+
"transformers_version": "5.6.0"
|
| 12 |
+
}
|
D41/checkpoint-5000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "<|begin_of_text|>",
|
| 4 |
+
"clean_up_tokenization_spaces": true,
|
| 5 |
+
"eos_token": "<|end_of_text|>",
|
| 6 |
+
"is_local": true,
|
| 7 |
+
"local_files_only": false,
|
| 8 |
+
"model_input_names": [
|
| 9 |
+
"input_ids",
|
| 10 |
+
"attention_mask"
|
| 11 |
+
],
|
| 12 |
+
"model_max_length": 131072,
|
| 13 |
+
"pad_token": "<|end_of_text|>",
|
| 14 |
+
"padding_side": "right",
|
| 15 |
+
"split_special_tokens": false,
|
| 16 |
+
"tokenizer_class": "TokenizersBackend"
|
| 17 |
+
}
|
D41/checkpoint-5000/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
D41/trainer_log.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
D41/viz_html/layer_01_mean_fixed_canvas_v3/assets/checkpoint-1000/fixed_canvas_v3_w420_h315_dpi140_layer_01_group_00_panel_00_mean.png
ADDED
|
D41/viz_html/layer_01_mean_fixed_canvas_v3/assets/checkpoint-1000/fixed_canvas_v3_w420_h315_dpi140_layer_01_group_00_panel_01_mean.png
ADDED
|
D41/viz_html/layer_01_mean_fixed_canvas_v3/assets/checkpoint-1000/fixed_canvas_v3_w420_h315_dpi140_layer_01_group_00_panel_02_mean.png
ADDED
|
D41/viz_html/layer_01_mean_fixed_canvas_v3/assets/checkpoint-1000/fixed_canvas_v3_w420_h315_dpi140_layer_01_group_01_panel_00_mean.png
ADDED
|
D41/viz_html/layer_01_mean_fixed_canvas_v3/assets/checkpoint-1000/fixed_canvas_v3_w420_h315_dpi140_layer_01_group_01_panel_01_mean.png
ADDED
|
D41/viz_html/layer_01_mean_fixed_canvas_v3/assets/checkpoint-4000/fixed_canvas_v3_w420_h315_dpi140_layer_01_group_00_panel_00_mean.png
ADDED
|
D41/viz_html/layer_01_mean_fixed_canvas_v3/assets/checkpoint-4000/fixed_canvas_v3_w420_h315_dpi140_layer_01_group_01_panel_01_mean.png
ADDED
|
D41/viz_html/layer_01_mean_fixed_canvas_v3/assets/checkpoint-5000/fixed_canvas_v3_w420_h315_dpi140_layer_01_group_00_panel_01_mean.png
ADDED
|
D63/checkpoint-1000/chat_template.jinja
ADDED
|
@@ -0,0 +1 @@
|
|
|
|
|
|
|
| 1 |
+
{% if messages[0]['role'] == 'system' %}{% set loop_messages = messages[1:] %}{% set system_message = messages[0]['content'] %}{% else %}{% set loop_messages = messages %}{% endif %}{% if system_message is defined %}{{ system_message }}{% endif %}{% for message in loop_messages %}{% set content = message['content'] %}{% if message['role'] == 'user' %}{{ content }}{% elif message['role'] == 'assistant' %}{{ content }}{% endif %}{% endfor %}
|
D63/checkpoint-1000/config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"architectures": [
|
| 3 |
+
"LlamaForCausalLM"
|
| 4 |
+
],
|
| 5 |
+
"attention_bias": false,
|
| 6 |
+
"attention_dropout": 0.0,
|
| 7 |
+
"bos_token_id": 128000,
|
| 8 |
+
"dtype": "float32",
|
| 9 |
+
"eos_token_id": 128001,
|
| 10 |
+
"head_dim": 64,
|
| 11 |
+
"hidden_act": "silu",
|
| 12 |
+
"hidden_size": 2048,
|
| 13 |
+
"initializer_range": 0.02,
|
| 14 |
+
"intermediate_size": 8192,
|
| 15 |
+
"max_position_embeddings": 131072,
|
| 16 |
+
"mlp_bias": false,
|
| 17 |
+
"model_type": "llama",
|
| 18 |
+
"num_attention_heads": 32,
|
| 19 |
+
"num_hidden_layers": 16,
|
| 20 |
+
"num_key_value_heads": 8,
|
| 21 |
+
"pad_token_id": 128001,
|
| 22 |
+
"pretraining_tp": 1,
|
| 23 |
+
"rms_norm_eps": 1e-05,
|
| 24 |
+
"rope_parameters": {
|
| 25 |
+
"factor": 32.0,
|
| 26 |
+
"high_freq_factor": 4.0,
|
| 27 |
+
"low_freq_factor": 1.0,
|
| 28 |
+
"original_max_position_embeddings": 8192,
|
| 29 |
+
"rope_theta": 500000.0,
|
| 30 |
+
"rope_type": "llama3"
|
| 31 |
+
},
|
| 32 |
+
"tie_word_embeddings": true,
|
| 33 |
+
"transformers_version": "5.6.0",
|
| 34 |
+
"use_cache": false,
|
| 35 |
+
"vocab_size": 128256
|
| 36 |
+
}
|
D63/checkpoint-1000/generation_config.json
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 128000,
|
| 4 |
+
"do_sample": true,
|
| 5 |
+
"eos_token_id": [
|
| 6 |
+
128001
|
| 7 |
+
],
|
| 8 |
+
"pad_token_id": 128001,
|
| 9 |
+
"temperature": 0.6,
|
| 10 |
+
"top_p": 0.9,
|
| 11 |
+
"transformers_version": "5.6.0"
|
| 12 |
+
}
|
D63/checkpoint-1000/tokenizer_config.json
ADDED
|
@@ -0,0 +1,17 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"backend": "tokenizers",
|
| 3 |
+
"bos_token": "<|begin_of_text|>",
|
| 4 |
+
"clean_up_tokenization_spaces": true,
|
| 5 |
+
"eos_token": "<|end_of_text|>",
|
| 6 |
+
"is_local": true,
|
| 7 |
+
"local_files_only": false,
|
| 8 |
+
"model_input_names": [
|
| 9 |
+
"input_ids",
|
| 10 |
+
"attention_mask"
|
| 11 |
+
],
|
| 12 |
+
"model_max_length": 131072,
|
| 13 |
+
"pad_token": "<|end_of_text|>",
|
| 14 |
+
"padding_side": "right",
|
| 15 |
+
"split_special_tokens": false,
|
| 16 |
+
"tokenizer_class": "TokenizersBackend"
|
| 17 |
+
}
|
D63/checkpoint-1000/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
D63/checkpoint-10000/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
D63/checkpoint-11000/trainer_state.json
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|