INDEX

Explanations

references to international summits and discussions

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

 quint

-0.15

AtA

-0.14

enumerator

-0.14

 Kendrick

-0.14

VML

-0.14

peq

-0.13

hk

-0.13

 Fowler

-0.13

ocoder

-0.13

çĵľ

-0.13

POSITIVE LOGITS

0.27

0.24

 summit

0.22

 Summit

0.21

0.20

 summ

0.20

0.19

LOB

0.19

-sum

0.19

lob

0.18

Activations Density 0.007%

references to international summits and discussions

No Comments

No Known Activations