INDEX

Explanations

references to royal titles or institutions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

CCR

-0.17

yb

-0.15

emit

-0.15

iola

-0.15

uss

-0.14

etal

-0.14

el

-0.14

agen

-0.14

Mutable

-0.14

POSITIVE LOGITS

izing

0.20

ilty

0.19

ization

0.18

izations

0.18

alty

0.18

isation

0.18

ised

0.17

zed

0.17

ized

0.16

ising

0.16

Activations Density 0.018%

references to royal titles or institutions

No Comments

No Known Activations