INDEX

Explanations

elements related to visual representations and their transformations

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

 ((__

-0.15

å¢ĥ

-0.14

\Context

-0.14

 VECTOR

-0.14

oug

-0.14

preferred

-0.14

oron

-0.14

Routing

-0.13

endor

-0.13

(æľĪ

-0.13

POSITIVE LOGITS

 pose

0.29

 poses

0.25

 Pose

0.23

 camera

0.23

 Stereo

0.23

 Bundle

0.23

 stereo

0.22

Ster

0.21

poses

0.21

Bundle

0.20

Activations Density 0.009%

elements related to visual representations and their transformations

No Comments

No Known Activations