INDEX

Explanations

phrases that indicate dependency or causation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

ibold

-0.16

Å¥

-0.16

Ñī

-0.15

Ø§ÙĦØ¥ÙĨØ¬ÙĦÙĬØ²ÙĬØ©

-0.15

undy

-0.15

STONE

-0.14

edly

-0.14

tro

-0.14

Sense

-0.14

tek

-0.14

POSITIVE LOGITS

ocks

0.15

0.14

veh

0.14

éĻµ

0.14

amburger

0.14

rette

0.14

Fallback

0.14

.compress

0.14

_servers

0.14

Activations Density 0.046%

phrases that indicate dependency or causation

No Comments

No Known Activations