d5b85a15fdba · 5.7GB
-
general.architectureolmoolmo
-
general.file_typeQ6_KQ6_K
-
olmo.attention.head_count3232
-
olmo.attention.head_count_kv3232
-
olmo.attention.layer_norm_epsilon1e-051e-05
-
olmo.block_count3232
-
olmo.context_length20482048
-
olmo.embedding_length40964096
-
olmo.feed_forward_length1100811008
-
olmo.rope.freq_base1000010000
-
tokenizer.ggml.add_bos_tokenfalsefalse
-
tokenizer.ggml.add_eos_tokenfalsefalse
-
tokenizer.ggml.eos_token_id5027950279
-
tokenizer.ggml.merges[Ġ Ġ, Ġ t, Ġ a, h e, i n, ...][Ġ Ġ, Ġ t, Ġ a, h e, i n, ...]
-
tokenizer.ggml.modelgpt2gpt2
-
tokenizer.ggml.padding_token_id11
-
tokenizer.ggml.preolmoolmo
-
tokenizer.ggml.token_type[4, 3, 1, 1, 1, ...][4, 3, 1, 1, 1, ...]
-
tokenizer.ggml.tokens[|||IP_ADDRESS|||, <|padding|>, !, ", #, ...][|||IP_ADDRESS|||, <|padding|>, !, ", #, ...]
-
token_embd.weightQ6_K[4096, 50304]
-
blk.0.attn_k.weightQ6_K[4096, 4096]
-
blk.0.attn_output.weightQ6_K[4096, 4096]
-
blk.0.attn_q.weightQ6_K[4096, 4096]
-
blk.0.attn_v.weightQ6_K[4096, 4096]
-
blk.0.ffn_down.weightQ6_K[11008, 4096]
-
blk.0.ffn_gate.weightQ6_K[4096, 11008]
-
blk.0.ffn_up.weightQ6_K[4096, 11008]
-
blk.1.attn_k.weightQ6_K[4096, 4096]
-
blk.1.attn_output.weightQ6_K[4096, 4096]
-
blk.1.attn_q.weightQ6_K[4096, 4096]
-
blk.1.attn_v.weightQ6_K[4096, 4096]
-
blk.1.ffn_down.weightQ6_K[11008, 4096]
-
blk.1.ffn_gate.weightQ6_K[4096, 11008]
-
blk.1.ffn_up.weightQ6_K[4096, 11008]
-
blk.2.attn_k.weightQ6_K[4096, 4096]
-
blk.2.attn_output.weightQ6_K[4096, 4096]
-
blk.2.attn_q.weightQ6_K[4096, 4096]
-
blk.2.attn_v.weightQ6_K[4096, 4096]
-
blk.2.ffn_down.weightQ6_K[11008, 4096]
-
blk.2.ffn_gate.weightQ6_K[4096, 11008]
-
blk.2.ffn_up.weightQ6_K[4096, 11008]
-
blk.3.attn_k.weightQ6_K[4096, 4096]
-
blk.3.attn_output.weightQ6_K[4096, 4096]
-
blk.3.attn_q.weightQ6_K[4096, 4096]
-
blk.3.attn_v.weightQ6_K[4096, 4096]
-
blk.3.ffn_down.weightQ6_K[11008, 4096]
-
blk.3.ffn_gate.weightQ6_K[4096, 11008]
-
blk.3.ffn_up.weightQ6_K[4096, 11008]
-
blk.4.attn_k.weightQ6_K[4096, 4096]
-
blk.4.attn_output.weightQ6_K[4096, 4096]
-
blk.4.attn_q.weightQ6_K[4096, 4096]
-
blk.4.attn_v.weightQ6_K[4096, 4096]
-
blk.4.ffn_down.weightQ6_K[11008, 4096]
-
blk.4.ffn_gate.weightQ6_K[4096, 11008]
-
blk.4.ffn_up.weightQ6_K[4096, 11008]
-
blk.5.attn_k.weightQ6_K[4096, 4096]
-
blk.5.attn_output.weightQ6_K[4096, 4096]
-
blk.5.attn_q.weightQ6_K[4096, 4096]
-
blk.5.attn_v.weightQ6_K[4096, 4096]
-
blk.5.ffn_down.weightQ6_K[11008, 4096]
-
blk.5.ffn_gate.weightQ6_K[4096, 11008]
-
blk.5.ffn_up.weightQ6_K[4096, 11008]
-
blk.6.attn_k.weightQ6_K[4096, 4096]
-
blk.6.attn_output.weightQ6_K[4096, 4096]
-
blk.6.attn_q.weightQ6_K[4096, 4096]
-
blk.6.attn_v.weightQ6_K[4096, 4096]
-
blk.6.ffn_down.weightQ6_K[11008, 4096]
-
blk.6.ffn_gate.weightQ6_K[4096, 11008]
-
blk.6.ffn_up.weightQ6_K[4096, 11008]
-
blk.7.attn_k.weightQ6_K[4096, 4096]
-
blk.7.attn_output.weightQ6_K[4096, 4096]
-
blk.7.attn_q.weightQ6_K[4096, 4096]
-
blk.7.attn_v.weightQ6_K[4096, 4096]
-
blk.7.ffn_down.weightQ6_K[11008, 4096]
-
blk.7.ffn_gate.weightQ6_K[4096, 11008]
-
blk.7.ffn_up.weightQ6_K[4096, 11008]
-
blk.8.attn_k.weightQ6_K[4096, 4096]
-
blk.8.attn_output.weightQ6_K[4096, 4096]
-
blk.8.attn_q.weightQ6_K[4096, 4096]
-
blk.8.attn_v.weightQ6_K[4096, 4096]
-
blk.8.ffn_down.weightQ6_K[11008, 4096]
-
blk.8.ffn_gate.weightQ6_K[4096, 11008]
-
blk.8.ffn_up.weightQ6_K[4096, 11008]
-
blk.9.attn_k.weightQ6_K[4096, 4096]
-
blk.9.attn_output.weightQ6_K[4096, 4096]
-
blk.9.attn_q.weightQ6_K[4096, 4096]
-
blk.9.attn_v.weightQ6_K[4096, 4096]
-
blk.9.ffn_down.weightQ6_K[11008, 4096]
-
blk.9.ffn_gate.weightQ6_K[4096, 11008]
-
blk.9.ffn_up.weightQ6_K[4096, 11008]
-
blk.10.attn_k.weightQ6_K[4096, 4096]
-
blk.10.attn_output.weightQ6_K[4096, 4096]
-
blk.10.attn_q.weightQ6_K[4096, 4096]
-
blk.10.attn_v.weightQ6_K[4096, 4096]
-
blk.10.ffn_down.weightQ6_K[11008, 4096]
-
blk.10.ffn_gate.weightQ6_K[4096, 11008]
-
blk.10.ffn_up.weightQ6_K[4096, 11008]
-
blk.11.attn_k.weightQ6_K[4096, 4096]
-
blk.11.attn_output.weightQ6_K[4096, 4096]
-
blk.11.attn_q.weightQ6_K[4096, 4096]
-
blk.11.attn_v.weightQ6_K[4096, 4096]
-
blk.11.ffn_down.weightQ6_K[11008, 4096]
-
blk.11.ffn_gate.weightQ6_K[4096, 11008]
-
blk.11.ffn_up.weightQ6_K[4096, 11008]
-
blk.12.attn_k.weightQ6_K[4096, 4096]
-
blk.12.attn_output.weightQ6_K[4096, 4096]
-
blk.12.attn_q.weightQ6_K[4096, 4096]
-
blk.12.attn_v.weightQ6_K[4096, 4096]
-
blk.12.ffn_down.weightQ6_K[11008, 4096]
-
blk.12.ffn_gate.weightQ6_K[4096, 11008]
-
blk.12.ffn_up.weightQ6_K[4096, 11008]
-
blk.13.attn_k.weightQ6_K[4096, 4096]
-
blk.13.attn_output.weightQ6_K[4096, 4096]
-
blk.13.attn_q.weightQ6_K[4096, 4096]
-
blk.13.attn_v.weightQ6_K[4096, 4096]
-
blk.13.ffn_down.weightQ6_K[11008, 4096]
-
blk.13.ffn_gate.weightQ6_K[4096, 11008]
-
blk.13.ffn_up.weightQ6_K[4096, 11008]
-
blk.14.attn_k.weightQ6_K[4096, 4096]
-
blk.14.attn_output.weightQ6_K[4096, 4096]
-
blk.14.attn_q.weightQ6_K[4096, 4096]
-
blk.14.attn_v.weightQ6_K[4096, 4096]
-
blk.14.ffn_down.weightQ6_K[11008, 4096]
-
blk.14.ffn_gate.weightQ6_K[4096, 11008]
-
blk.14.ffn_up.weightQ6_K[4096, 11008]
-
blk.15.attn_k.weightQ6_K[4096, 4096]
-
blk.15.attn_output.weightQ6_K[4096, 4096]
-
blk.15.attn_q.weightQ6_K[4096, 4096]
-
blk.15.attn_v.weightQ6_K[4096, 4096]
-
blk.15.ffn_down.weightQ6_K[11008, 4096]
-
blk.15.ffn_gate.weightQ6_K[4096, 11008]
-
blk.15.ffn_up.weightQ6_K[4096, 11008]
-
blk.16.attn_k.weightQ6_K[4096, 4096]
-
blk.16.attn_output.weightQ6_K[4096, 4096]
-
blk.16.attn_q.weightQ6_K[4096, 4096]
-
blk.16.attn_v.weightQ6_K[4096, 4096]
-
blk.16.ffn_down.weightQ6_K[11008, 4096]
-
blk.16.ffn_gate.weightQ6_K[4096, 11008]
-
blk.16.ffn_up.weightQ6_K[4096, 11008]
-
blk.17.attn_k.weightQ6_K[4096, 4096]
-
blk.17.attn_output.weightQ6_K[4096, 4096]
-
blk.17.attn_q.weightQ6_K[4096, 4096]
-
blk.17.attn_v.weightQ6_K[4096, 4096]
-
blk.17.ffn_down.weightQ6_K[11008, 4096]
-
blk.17.ffn_gate.weightQ6_K[4096, 11008]
-
blk.17.ffn_up.weightQ6_K[4096, 11008]
-
blk.18.attn_k.weightQ6_K[4096, 4096]
-
blk.18.attn_output.weightQ6_K[4096, 4096]
-
blk.18.attn_q.weightQ6_K[4096, 4096]
-
blk.18.attn_v.weightQ6_K[4096, 4096]
-
blk.18.ffn_down.weightQ6_K[11008, 4096]
-
blk.18.ffn_gate.weightQ6_K[4096, 11008]
-
blk.18.ffn_up.weightQ6_K[4096, 11008]
-
blk.19.attn_k.weightQ6_K[4096, 4096]
-
blk.19.attn_output.weightQ6_K[4096, 4096]
-
blk.19.attn_q.weightQ6_K[4096, 4096]
-
blk.19.attn_v.weightQ6_K[4096, 4096]
-
blk.19.ffn_down.weightQ6_K[11008, 4096]
-
blk.19.ffn_gate.weightQ6_K[4096, 11008]
-
blk.19.ffn_up.weightQ6_K[4096, 11008]
-
blk.20.attn_k.weightQ6_K[4096, 4096]
-
blk.20.attn_output.weightQ6_K[4096, 4096]
-
blk.20.attn_q.weightQ6_K[4096, 4096]
-
blk.20.attn_v.weightQ6_K[4096, 4096]
-
blk.20.ffn_down.weightQ6_K[11008, 4096]
-
blk.20.ffn_gate.weightQ6_K[4096, 11008]
-
blk.20.ffn_up.weightQ6_K[4096, 11008]
-
blk.21.attn_k.weightQ6_K[4096, 4096]
-
blk.21.attn_output.weightQ6_K[4096, 4096]
-
blk.21.attn_q.weightQ6_K[4096, 4096]
-
blk.21.attn_v.weightQ6_K[4096, 4096]
-
blk.21.ffn_down.weightQ6_K[11008, 4096]
-
blk.21.ffn_gate.weightQ6_K[4096, 11008]
-
blk.21.ffn_up.weightQ6_K[4096, 11008]
-
blk.22.attn_k.weightQ6_K[4096, 4096]
-
blk.22.attn_output.weightQ6_K[4096, 4096]
-
blk.22.attn_q.weightQ6_K[4096, 4096]
-
blk.22.attn_v.weightQ6_K[4096, 4096]
-
blk.22.ffn_down.weightQ6_K[11008, 4096]
-
blk.22.ffn_gate.weightQ6_K[4096, 11008]
-
blk.22.ffn_up.weightQ6_K[4096, 11008]
-
blk.23.attn_k.weightQ6_K[4096, 4096]
-
blk.23.attn_output.weightQ6_K[4096, 4096]
-
blk.23.attn_q.weightQ6_K[4096, 4096]
-
blk.23.attn_v.weightQ6_K[4096, 4096]
-
blk.23.ffn_down.weightQ6_K[11008, 4096]
-
blk.23.ffn_gate.weightQ6_K[4096, 11008]
-
blk.23.ffn_up.weightQ6_K[4096, 11008]
-
blk.24.attn_k.weightQ6_K[4096, 4096]
-
blk.24.attn_output.weightQ6_K[4096, 4096]
-
blk.24.attn_q.weightQ6_K[4096, 4096]
-
blk.24.attn_v.weightQ6_K[4096, 4096]
-
blk.24.ffn_down.weightQ6_K[11008, 4096]
-
blk.24.ffn_gate.weightQ6_K[4096, 11008]
-
blk.24.ffn_up.weightQ6_K[4096, 11008]
-
blk.25.attn_k.weightQ6_K[4096, 4096]
-
blk.25.attn_output.weightQ6_K[4096, 4096]
-
blk.25.attn_q.weightQ6_K[4096, 4096]
-
blk.25.attn_v.weightQ6_K[4096, 4096]
-
blk.25.ffn_down.weightQ6_K[11008, 4096]
-
blk.25.ffn_gate.weightQ6_K[4096, 11008]
-
blk.25.ffn_up.weightQ6_K[4096, 11008]
-
blk.26.attn_k.weightQ6_K[4096, 4096]
-
blk.26.attn_output.weightQ6_K[4096, 4096]
-
blk.26.attn_q.weightQ6_K[4096, 4096]
-
blk.26.attn_v.weightQ6_K[4096, 4096]
-
blk.26.ffn_down.weightQ6_K[11008, 4096]
-
blk.26.ffn_gate.weightQ6_K[4096, 11008]
-
blk.26.ffn_up.weightQ6_K[4096, 11008]
-
blk.27.attn_k.weightQ6_K[4096, 4096]
-
blk.27.attn_output.weightQ6_K[4096, 4096]
-
blk.27.attn_q.weightQ6_K[4096, 4096]
-
blk.27.attn_v.weightQ6_K[4096, 4096]
-
blk.27.ffn_down.weightQ6_K[11008, 4096]
-
blk.27.ffn_gate.weightQ6_K[4096, 11008]
-
blk.27.ffn_up.weightQ6_K[4096, 11008]
-
blk.28.attn_k.weightQ6_K[4096, 4096]
-
blk.28.attn_output.weightQ6_K[4096, 4096]
-
blk.28.attn_q.weightQ6_K[4096, 4096]
-
blk.28.attn_v.weightQ6_K[4096, 4096]
-
blk.28.ffn_down.weightQ6_K[11008, 4096]
-
blk.28.ffn_gate.weightQ6_K[4096, 11008]
-
blk.28.ffn_up.weightQ6_K[4096, 11008]
-
blk.29.attn_k.weightQ6_K[4096, 4096]
-
blk.29.attn_output.weightQ6_K[4096, 4096]
-
blk.29.attn_q.weightQ6_K[4096, 4096]
-
blk.29.attn_v.weightQ6_K[4096, 4096]
-
blk.29.ffn_down.weightQ6_K[11008, 4096]
-
blk.29.ffn_gate.weightQ6_K[4096, 11008]
-
blk.29.ffn_up.weightQ6_K[4096, 11008]
-
blk.30.attn_k.weightQ6_K[4096, 4096]
-
blk.30.attn_output.weightQ6_K[4096, 4096]
-
blk.30.attn_q.weightQ6_K[4096, 4096]
-
blk.30.attn_v.weightQ6_K[4096, 4096]
-
blk.30.ffn_down.weightQ6_K[11008, 4096]
-
blk.30.ffn_gate.weightQ6_K[4096, 11008]
-
blk.30.ffn_up.weightQ6_K[4096, 11008]
-
blk.31.attn_k.weightQ6_K[4096, 4096]
-
blk.31.attn_output.weightQ6_K[4096, 4096]
-
blk.31.attn_q.weightQ6_K[4096, 4096]
-
blk.31.attn_v.weightQ6_K[4096, 4096]
-
blk.31.ffn_down.weightQ6_K[11008, 4096]
-
blk.31.ffn_gate.weightQ6_K[4096, 11008]
-
blk.31.ffn_up.weightQ6_K[4096, 11008]
-
output.weightQ6_K[4096, 50304]
Metadata
Tensor
blk.0
blk.1
blk.2
blk.3
blk.4
blk.5
blk.6
blk.7
blk.8
blk.9
blk.10
blk.11
blk.12
blk.13
blk.14
blk.15
blk.16
blk.17
blk.18
blk.19
blk.20
blk.21
blk.22
blk.23
blk.24
blk.25
blk.26
blk.27
blk.28
blk.29
blk.30
blk.31