INDEX

Explanations

phrases indicating support or affirmation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

 Rouge

-0.16

anch

-0.15

DM

-0.15

eka

-0.14

UG

-0.14

 conj

-0.14

ATS

-0.14

 favor

-0.14

ad

-0.14

eness

-0.14

POSITIVE LOGITS

 backing

0.32

Backing

0.27

/back

0.24

 backed

0.23

 backs

0.21

-backed

0.21

(back

0.20

=back

0.20

haul

0.19

aret

0.18

Activations Density 0.015%

phrases indicating support or affirmation

No Comments

No Known Activations