INDEX

Explanations

actions related to climbing or ascending

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

alley

-0.18

 downward

-0.17

arme

-0.15

éľĩ

-0.15

ingleton

-0.14

nero

-0.14

 IonicPage

-0.14

rina

-0.14

xing

-0.14

HB

-0.14

POSITIVE LOGITS

 Mount

0.22

Mt

0.21

 stairs

0.21

 aboard

0.20

 ladder

0.19

 onto

0.19

Jacob

0.18

 Jacob

0.18

Mount

0.18

Kil

0.18

Activations Density 0.017%

actions related to climbing or ascending

No Comments

No Known Activations