INDEX

Explanations

phrases related to changes in states or conditions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ahn

-0.16

 satur

-0.15

oras

-0.14

erken

-0.14

ach

-0.14

 SqlDbType

-0.13

reff

-0.13

amer

-0.13

POSITIVE LOGITS

 again

0.33

again

0.30

 Again

0.26

Again

0.24

 novamente

0.24

 láº¡i

0.23

 ÑģÐ½Ð¾Ð²Ð°

0.22

_again

0.22

 ëĭ¤ìĭľ

0.21

 wieder

0.21

Activations Density 0.078%

phrases related to changes in states or conditions

No Comments

No Known Activations