INDEX

Explanations

phrases related to comparisons and rankings

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

 wards

-0.18

-0.16

akest

-0.15

uluk

-0.15

pta

-0.15

udios

-0.15

beck

-0.15

upe

-0.15

agma

-0.14

 ward

-0.14

POSITIVE LOGITS

 Pere

0.15

Î½Î¿Î¹

0.15

ÏģÎ¿Ïį

0.15

 among

0.14

 Parl

0.14

æĹģ

0.14

uran

0.14

_locals

0.14

aily

0.14

mgr

0.14

Activations Density 0.435%

phrases related to comparisons and rankings

No Comments

No Known Activations