INDEX

Explanations

the name "Mulligan" or variations of it associated with a legal context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

olest

-0.20

angler

-0.18

ovol

-0.18

licate

-0.17

osy

-0.17

eff

-0.16

oyo

-0.16

other

-0.16

ole

-0.16

emit

-0.16

POSITIVE LOGITS

er

0.25

0.23

igan

0.22

ered

0.22

erb

0.21

0.20

eri

0.19

ery

0.19

ervo

0.18

iams

0.17

Activations Density 0.014%

the name "Mulligan" or variations of it associated with a legal context

No Comments

No Known Activations