INDEX

Explanations

elements that emphasize descriptive qualities in writing

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

ABCDEFGHI

-0.15

anitize

-0.15

XL

-0.15

 completion

-0.14

 hoax

-0.14

 Completion

-0.14

 Gale

-0.13

completion

-0.13

 serial

-0.13

ABCDEFGHIJKLMNOP

-0.13

POSITIVE LOGITS

 description

0.33

 descriptions

0.31

 dialogue

0.28

 Description

0.27

-dialog

0.25

description

0.25

Dialogue

0.25

 descriptive

0.24

 Dialogue

0.24

_description

0.24

Activations Density 0.113%

elements that emphasize descriptive qualities in writing

No Comments

No Known Activations