INDEX

Explanations

occurrences of the word "up" or related derivatives

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

-0.19

quip

-0.18

blem

-0.15

tail

-0.15

arp

-0.15

eph

-0.15

ament

-0.14

ÃŃsto

-0.14

auer

-0.14

usiness

-0.14

POSITIVE LOGITS

/down

0.23

ILON

0.19

-down

0.18

yun

0.18

otre

0.17

ended

0.17

ozilla

0.16

ToDate

0.16

adesh

0.15

bra

0.15

Activations Density 0.039%

occurrences of the word "up" or related derivatives

No Comments

No Known Activations