INDEX

Explanations

phrases indicating belonging or membership within a group

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

oun

-0.16

.scalablytyped

-0.15

osed

-0.15

obia

-0.14

hoff

-0.14

kÄ±r

-0.14

uler

-0.14

zas

-0.14

Ã¼nd

-0.14

POSITIVE LOGITS

few

0.20

Few

0.16

ema

0.15

maal

0.15

strup

0.15

 many

0.15

åĮ

0.14

apo

0.14

_many

0.14

 fier

0.14

Activations Density 0.120%

phrases indicating belonging or membership within a group

No Comments

No Known Activations