INDEX

Explanations

phrases with a strong emphasis on adjectives and descriptors related to quality

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 28-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L28R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.28.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

ì§ĵ

-0.16

.si

-0.14

ÑģÑĤÐ¸

-0.14

Ã©rica

-0.14

ava

-0.14

licity

-0.14

eras

-0.14

å¾¡

-0.14

inals

-0.13

inalg

-0.13

POSITIVE LOGITS

 tale

0.23

 departure

0.20

ode

0.18

 celebration

0.18

 caution

0.18

pa

0.18

 tour

0.18

 bild

0.18

 glimpse

0.17

 feast

0.17

Activations Density 0.095%

phrases with a strong emphasis on adjectives and descriptors related to quality

No Comments

No Known Activations