INDEX

Explanations

terms related to solubility

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

ess

-0.21

plib

-0.19

Ä±rak

-0.18

esc

-0.17

estro

-0.17

plural

-0.16

inning

-0.16

dro

-0.15

ee

-0.15

yll

-0.15

POSITIVE LOGITS

omon

0.28

uble

0.25

vable

0.24

itaire

0.23

stice

0.23

vents

0.22

licit

0.22

itude

0.20

stitial

0.19

sol

0.19

Activations Density 0.017%

terms related to solubility

No Comments

No Known Activations