INDEX

Explanations

forms of the verb "to be."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

ton

-0.18

aro

-0.15

ãĥ¼ãĤ

-0.15

tons

-0.14

gence

-0.14

holm

-0.14

enko

-0.14

ucht

-0.14

esse

-0.14

itical

-0.13

POSITIVE LOGITS

/is

0.24

/how

0.24

/w

0.23

 supposed

0.21

/do

0.20

 wont

0.17

 rumored

0.16

±Ð¾ÑĤ

0.15

 gonna

0.15

 prÅ¯bÄĽhu

0.15

Activations Density 0.110%

forms of the verb "to be."

No Comments

No Known Activations