INDEX

Explanations

technical terms and code snippets related to programming functions and algorithms

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

Ð²ÑĸÑĤ

-0.17

pek

-0.17

 trÃ¬

-0.14

Ã²n

-0.14

arak

-0.14

 Detail

-0.13

_iff

-0.13

Charts

-0.13

wan

-0.13

 drag

-0.13

POSITIVE LOGITS

np

0.40

np

0.34

NP

0.32

(np

0.32

=np

0.29

[:,

0.25

NP

0.25

 mask

0.24

[np

0.23

 Mask

0.23

Activations Density 0.030%

technical terms and code snippets related to programming functions and algorithms

No Comments

No Known Activations