INDEX

Explanations

adverbs indicating careful or precise actions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

 have

-0.19

has

-0.19

 carefully

-0.19

 being

-0.18

is

-0.18

was

-0.17

 successfully

-0.17

 prominently

-0.17

 having

-0.17

 meant

-0.16

POSITIVE LOGITS

and

0.36

yet

0.34

yet

0.28

but

0.28

Yet

0.25

-but

0.23

Yet

0.23

 enough

0.22

but

0.21

and

0.21

Activations Density 0.112%

adverbs indicating careful or precise actions

No Comments

No Known Activations