INDEX

Explanations

phrases indicating a need or necessity

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

etto

-0.17

á»Ń

-0.17

-0.16

ainter

-0.15

 æĶ¶

-0.15

orbit

-0.15

Î»Î¹Îº

-0.14

nant

-0.14

ermo

-0.14

pu

-0.14

POSITIVE LOGITS

uro

0.16

asd

0.15

Ler

0.15

opi

0.14

leet

0.14

-tm

0.14

upp

0.14

isd

0.14

sov

0.13

 Alley

0.13

Activations Density 0.077%

phrases indicating a need or necessity

No Comments

No Known Activations