INDEX

Explanations

occurrences of the letter "W."

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

ebra

-0.18

.struts

-0.17

IVA

-0.17

own

-0.17

ivas

-0.15

ow

-0.15

ribly

-0.15

owards

-0.15

zig

-0.15

earer

-0.15

POSITIVE LOGITS

HEEL

0.23

ALK

0.22

ORLD

0.21

OOD

0.21

ORTH

0.20

ELCOME

0.20

ILD

0.19

ASHINGTON

0.19

ASTE

0.18

ARS

0.18

Activations Density 0.013%

occurrences of the letter "W."

No Comments

No Known Activations