INDEX

Explanations

terms related to walking or walk-in features in spaces

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

riers

-0.18

ãĥ¼ãĥĭ

-0.16

inx

-0.15

 bÃ½

-0.15

udden

-0.15

á»ħ

-0.15

lte

-0.14

illos

-0.14

iang

-0.14

ilation

-0.14

POSITIVE LOGITS

way

0.22

Ã¼re

0.20

 Walk

0.19

adoo

0.18

 walk

0.18

WAY

0.18

Walk

0.17

ways

0.17

walk

0.15

ERA

0.15

Activations Density 0.012%

terms related to walking or walk-in features in spaces

No Comments

No Known Activations