INDEX

Explanations

occurrences of the letter "T" in various forms and contexts

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

pector

-0.20

Ìģc

-0.17

inator

-0.16

THON

-0.15

lettes

-0.15

lica

-0.15

 Hlav

-0.15

Ã¼ltÃ¼r

-0.14

leton

-0.14

leston

-0.14

POSITIVE LOGITS

ats

0.31

hey

0.23

his

0.23

ere

0.22

ogh

0.22

eres

0.21

rowing

0.21

ough

0.21

ose

0.21

Is

0.21

Activations Density 0.013%

occurrences of the letter "T" in various forms and contexts

No Comments

No Known Activations