INDEX

Explanations

repetitive or common phrases used in instructions or troubleshooting

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

clamp

-0.16

eniable

-0.15

 honor

-0.14

 ÙĤØ³Ùħ

-0.14

 redraw

-0.14

 Recycling

-0.14

 Mature

-0.14

 redesigned

-0.14

/buttons

-0.13

/ref

-0.13

POSITIVE LOGITS

run

0.23

 import

0.23

 install

0.22

 launch

0.20

 Import

0.20

 lunch

0.20

 Install

0.19

run

0.19

 deploy

0.19

å¯¼

0.19

Activations Density 0.156%

repetitive or common phrases used in instructions or troubleshooting

No Comments

No Known Activations