INDEX

Explanations

references to "super" with various contexts, indicating a focus on extraordinary qualities or characteristics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

sj

-0.25

sWith

-0.23

sb

-0.22

sel

-0.22

sch

-0.22

sworth

-0.21

-0.20

side

-0.20

sx

-0.20

sz

-0.19

POSITIVE LOGITS

iors

0.38

charged

0.38

stit

0.35

lat

0.31

powers

0.30

cil

0.30

iore

0.29

human

0.29

stars

0.29

charger

0.29

Activations Density 0.027%

references to "super" with various contexts, indicating a focus on extraordinary qualities or characteristics

No Comments

No Known Activations