INDEX

Explanations

phrases indicating intentions or goals involving actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

venue

-0.17

ingly

-0.16

endar

-0.15

alam

-0.15

 âĹĦ

-0.15

_marshall

-0.15

ots

-0.15

owers

-0.14

Ã¤ng

-0.14

edly

-0.14

POSITIVE LOGITS

evin

0.17

ĶĦ

0.14

,'#

0.14

_gradient

0.13

Deb

0.13

ÐµÑı

0.13

$MESS

0.13

¢

0.13

dbus

0.13

azen

0.13

Activations Density 0.535%

phrases indicating intentions or goals involving actions

No Comments

No Known Activations