INDEX

Explanations

occurrences of the word "a" and its variations, highlighting definite articles and nouns

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

hek

-0.21

ftype

-0.17

ãĥ«ãĥķ

-0.15

orphic

-0.15

asio

-0.15

ebra

-0.15

ÄĽt

-0.15

lil

-0.14

cents

-0.14

fault

-0.14

POSITIVE LOGITS

venture

0.30

ube

0.29

igu

0.29

Ã©ro

0.28

quit

0.25

ig

0.25

oste

0.25

vertis

0.25

ute

0.24

uteur

0.24

Activations Density 0.015%

occurrences of the word "a" and its variations, highlighting definite articles and nouns

No Comments

No Known Activations