INDEX

Explanations

terms related to wet and dry conditions, particularly in the context of environmental descriptions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

opers

-0.19

mekte

-0.17

ials

-0.17

azi

-0.16

yx

-0.15

uy

-0.15

ially

-0.15

ots

-0.15

oted

-0.15

erable

-0.14

POSITIVE LOGITS

ted

0.31

ting

0.30

lands

0.28

zel

0.28

ter

0.27

ters

0.26

ten

0.24

test

0.23

ensch

0.21

tring

0.21

Activations Density 0.010%

terms related to wet and dry conditions, particularly in the context of environmental descriptions

No Comments

No Known Activations