INDEX

Explanations

phrases related to information access or accessibility

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 17-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L17R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.17.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

ationToken

-0.17

rena

-0.16

ly

-0.15

agan

-0.15

isha

-0.14

ishly

-0.14

icken

-0.14

ellas

-0.14

 lÃłnh

-0.14

ickey

-0.13

POSITIVE LOGITS

ibly

0.33

ions

0.26

ibilities

0.26

ibility

0.24

ories

0.24

ible

0.22

orial

0.22

ibil

0.22

ability

0.22

ory

0.21

Activations Density 0.038%

phrases related to information access or accessibility

No Comments

No Known Activations