INDEX

Explanations

occurrences of the word "on."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

bage

-0.18

worth

-0.15

 regards

-0.15

irk

-0.15

Ã¢k

-0.15

vester

-0.15

fighter

-0.15

 regard

-0.14

gether

-0.14

ories

-0.14

POSITIVE LOGITS

yx

0.22

look

0.19

coming

0.18

omat

0.18

gin

0.18

ymous

0.17

este

0.17

ething

0.16

ep

0.15

/off

0.15

Activations Density 0.065%

occurrences of the word "on."

No Comments

No Known Activations