INDEX

Explanations

references to friends and social connections

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

eria

-0.18

OPLE

-0.18

arin

-0.17

orgot

-0.17

ft

-0.17

undi

-0.16

erras

-0.16

erse

-0.16

itz

-0.16

baz

-0.15

POSITIVE LOGITS

liness

0.34

lier

0.34

liest

0.32

ships

0.31

ship

0.25

/ac

0.25

lies

0.25

shipping

0.22

/f

0.20

SHIP

0.20

Activations Density 0.054%

references to friends and social connections

No Comments

No Known Activations