INDEX

Explanations

activities and conditions related to physical labor and injury, particularly those common in occupational settings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

stance

-0.15

zen

-0.15

olated

-0.15

lier

-0.15

xon

-0.14

Ð¾ÑĢÐ¾Ð¶

-0.14

coe

-0.14

orage

-0.14

ãĥªãĤ¢

-0.14

ophe

-0.13

POSITIVE LOGITS

 exposures

0.24

 exposure

0.23

 Exposure

0.21

è·Ŀ

0.18

Exposed

0.18

 unnatural

0.18

unsupported

0.15

 exposed

0.15

Äħd

0.15

Expose

0.15

Activations Density 0.155%

activities and conditions related to physical labor and injury, particularly those common in occupational settings

No Comments

No Known Activations