INDEX

Explanations

phrases indicating a minimum amount or threshold

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

 even

-0.18

 EVEN

-0.17

 actually

-0.16

aphore

-0.16

even

-0.16

anio

-0.16

çĶļèĩ³

-0.15

 ÑħÐ¾ÑĤÑı

-0.15

 either

-0.15

uchen

-0.15

POSITIVE LOGITS

 until

0.22

 according

0.21

until

0.20

Until

0.19

 ones

0.18

 Until

0.18

according

0.16

asm

0.15

 According

0.15

 hasta

0.15

Activations Density 0.025%

phrases indicating a minimum amount or threshold

No Comments

No Known Activations