INDEX

Explanations

conjunctions and phrases indicating condition or relation

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

-generic

-0.17

 generic

-0.17

else

-0.16

_generic

-0.15

Generic

-0.15

Dou

-0.15

yiy

-0.15

rud

-0.15

 else

-0.14

nicas

-0.14

POSITIVE LOGITS

 whatever

0.32

whatever

0.30

 anything

0.26

 something

0.24

Whatever

0.23

something

0.23

 Whatever

0.22

anything

0.21

 Anything

0.21

Anything

0.20

Activations Density 0.008%

conjunctions and phrases indicating condition or relation

No Comments

No Known Activations