INDEX

Explanations

elements related to navigation and object manipulation within robotics contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

DNA

-0.14

grade

-0.14

adil

-0.14

GRADE

-0.13

Bur

-0.13

lices

-0.13

 åħ

-0.13

 Grade

-0.13

ascending

-0.13

POSITIVE LOGITS

 scene

0.23

oc

0.22

 scenes

0.21

 object

0.21

Oc

0.19

scene

0.19

 objects

0.19

 Scene

0.19

 Scenes

0.18

objects

0.18

Activations Density 0.035%

elements related to navigation and object manipulation within robotics contexts

No Comments

No Known Activations