INDEX

Explanations

negations or phrases expressing the absence or rejection of something

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ans

-0.17

lings

-0.14

gow

-0.14

opot

-0.14

cul

-0.14

tra

-0.14

encers

-0.14

findAll

-0.14

 Ð²ÑģÐµÐ³Ð¾

-0.14

ndon

-0.13

POSITIVE LOGITS

 amount

0.31

 matter

0.26

amount

0.25

 Amount

0.22

 aspect

0.21

one

0.21

 single

0.20

 doubt

0.19

things

0.19

Amount

0.19

Activations Density 0.046%

negations or phrases expressing the absence or rejection of something

No Comments

No Known Activations