INDEX

Explanations

phrases indicating inclusivity or diversity across different groups or categories

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

ignum

-0.17

orang

-0.16

XObject

-0.16

antu

-0.15

.hashCode

-0.14

lsi

-0.14

wort

-0.14

½æķ°

-0.14

 Either

-0.14

abilit

-0.13

POSITIVE LOGITS

 ages

0.17

kers

0.16

acles

0.16

age

0.16

Cly

0.14

 bÃ¼y

0.14

 size

0.14

å¤§å°ı

0.14

uen

0.13

ensch

0.13

Activations Density 0.040%

phrases indicating inclusivity or diversity across different groups or categories

No Comments

No Known Activations