INDEX

Explanations

the word "am" and its variants

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

gnore

-0.16

inize

-0.16

illac

-0.16

lust

-0.16

lus

-0.15

quin

-0.15

enheim

-0.15

geois

-0.15

agate

-0.15

APSHOT

-0.15

POSITIVE LOGITS

ends

0.28

putation

0.26

enable

0.25

iable

0.25

icable

0.25

orph

0.24

ending

0.24

assing

0.23

ulet

0.23

/pm

0.23

Activations Density 0.023%

the word "am" and its variants

No Comments

No Known Activations