INDEX

Explanations

expressions related to time or comparisons across years

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

hani

-0.15

ab

-0.15

 anti

-0.14

Ã½t

-0.14

unw

-0.14

Sco

-0.14

uche

-0.14

_timeout

-0.14

cmc

-0.14

otti

-0.14

POSITIVE LOGITS

-on

0.22

-over

0.19

ago

0.17

 growth

0.16

growth

0.16

-On

0.16

 Growth

0.15

ually

0.15

agma

0.14

protocols

0.14

Activations Density 0.010%

expressions related to time or comparisons across years

No Comments

No Known Activations