INDEX

Explanations

words related to bumping or impacting

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

undo

-0.15

park

-0.15

UTION

-0.15

RY

-0.15

onne

-0.15

ipped

-0.15

hn

-0.15

Ø±

-0.14

iser

-0.14

ë´ī

-0.14

POSITIVE LOGITS

sters

0.25

ster

0.22

kins

0.21

site

0.20

stead

0.20

ylon

0.19

urge

0.19

ers

0.18

shaft

0.18

ing

0.17

Activations Density 0.059%

words related to bumping or impacting

No Comments

No Known Activations