INDEX

Explanations

occurrences of phrases indicating significant impacts or consequences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

abr

-0.15

åĮ

-0.15

 nackte

-0.15

ething

-0.15

ecast

-0.14

polator

-0.14

\Bundle

-0.13

okud

-0.13

 Ð·Ð°Ð¼

-0.13

kest

-0.13

POSITIVE LOGITS

 pretty

0.21

bit

0.20

 welcome

0.19

 hell

0.18

fir

0.18

mod

0.17

 fitting

0.17

 mighty

0.17

bit

0.17

 decided

0.17

Activations Density 0.152%

occurrences of phrases indicating significant impacts or consequences

No Comments

No Known Activations