INDEX

Explanations

occurrences of the word "well."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ynes

-0.22

ollo

-0.16

igs

-0.16

yne

-0.15

inz

-0.14

ega

-0.14

linger

-0.14

ãĥ¼ãĥľ

-0.14

><?

-0.14

ivos

-0.14

POSITIVE LOGITS

as

0.36

-known

0.20

-being

0.18

 well

0.18

spring

0.17

 ÙĥÙħØ§

0.17

being

0.16

 ÐºÐ°Ðº

0.16

/or

0.15

 addition

0.15

Activations Density 0.030%

occurrences of the word "well."

No Comments

No Known Activations