INDEX

Explanations

references to the Himalayas

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

sie

-0.15

-bo

-0.14

hai

-0.14

imap

-0.14

erer

-0.14

↵

-0.14

 Mant

-0.14

 ladder

-0.14

 Monaco

-0.14

-0.13

POSITIVE LOGITS

ancode

0.16

zkum

0.15

oice

0.15

urd

0.15

appiness

0.15

arten

0.15

uyo

0.14

PubMed

0.14

 duct

0.14

SKI

0.14

Activations Density 0.005%

references to the Himalayas

No Comments

No Known Activations