INDEX

Explanations

instances of the word "similar" as a descriptor or comparison

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

een

-0.17

yp

-0.17

vÃŃ

-0.16

eln

-0.16

essa

-0.15

printStats

-0.15

eer

-0.15

eter

-0.14

ngr

-0.14

 hete

-0.14

POSITIVE LOGITS

-minded

0.23

ily

0.22

mente

0.21

-sex

0.20

teenth

0.18

weise

0.18

inded

0.17

 minded

0.17

etto

0.17

-looking

0.17

Activations Density 0.029%

instances of the word "similar" as a descriptor or comparison

No Comments

No Known Activations