INDEX

Explanations

phrases indicating excessive quantities or amounts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

 toch

-0.20

st

-0.18

wood

-0.17

 nÃło

-0.17

ry

-0.17

-0.16

pu

-0.16

core

-0.16

wy

-0.16

kar

-0.16

POSITIVE LOGITS

led

0.28

ledo

0.23

 boot

0.22

gether

0.21

oooooooo

0.19

/from

0.18

 Boot

0.18

thers

0.17

oooo

0.17

Activations Density 0.021%

phrases indicating excessive quantities or amounts

No Comments

No Known Activations