INDEX

Explanations

intensifiers that amplify characteristics or qualities

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

istically

-0.21

izing

-0.18

ishly

-0.18

ensely

-0.17

etically

-0.17

ising

-0.16

å°ĳ

-0.16

bor

-0.16

ologically

-0.15

 like

-0.15

POSITIVE LOGITS

ÑĪÐºÐ°

0.17

FAULT

0.16

upp

0.14

eniable

0.14

AndGet

0.14

-Ð³

0.14

 Kraj

0.13

 Advantage

0.13

uppy

0.13

nech

0.13

Activations Density 0.088%

intensifiers that amplify characteristics or qualities

No Comments

No Known Activations