INDEX

Explanations

references to specific individuals involved in relationship dynamics and conflicts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

okable

-0.16

rawer

-0.16

subst

-0.15

uby

-0.15

ëģĶ

-0.15

ideo

-0.14

ldb

-0.14

orthand

-0.14

 vitam

-0.14

veral

-0.13

POSITIVE LOGITS

has

0.23

may

0.23

is

0.22

 shouldn

0.21

 deserves

0.20

 should

0.20

 certainly

0.20

isn

0.20

may

0.20

 might

0.19

Activations Density 0.686%

references to specific individuals involved in relationship dynamics and conflicts

No Comments

No Known Activations