INDEX

Explanations

negations or phrases indicating the absence 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 @ 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

rix

-0.16

_UNUSED

-0.15

 subur

-0.15

agle

-0.15

Characteristic

-0.15

PIO

-0.14

 inne

-0.14

éĮ²

-0.14

/star

-0.14

 ÑĥÑģÐµ

-0.14

POSITIVE LOGITS

 conventional

0.17

 traditional

0.16

traditional

0.15

wald

0.14

ori

0.14

 surprise

0.14

uala

0.14

Traditional

0.14

 convention

0.14

 Traditional

0.14

Activations Density 0.391%

negations or phrases indicating the absence of something

No Comments

No Known Activations