INDEX

Explanations

actions involving initiating, banning, or shaping significant events and behaviors

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

able

-0.17

ordo

-0.15

ÏħÏĦÏĮ

-0.15

alley

-0.15

ãĤĤãĤĬ

-0.14

/by

-0.14

ÑĤÐ°Ð»

-0.14

Ã¨

-0.14

tet

-0.14

zzo

-0.13

POSITIVE LOGITS

/is

0.22

nt

0.20

iate

0.19

 Ð±Ñĭ

0.19

phin

0.19

ness

0.18

rale

0.18

leine

0.15

iates

0.15

actic

0.15

Activations Density 1.071%

actions involving initiating, banning, or shaping significant events and behaviors

No Comments

No Known Activations