INDEX

Explanations

phrases indicating the development or existence of programs and services

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

ãĥĵãĥ¼

-0.17

phy

-0.15

uada

-0.15

otime

-0.15

orm

-0.15

 experimentation

-0.14

 Desc

-0.14

 experiments

-0.14

ipl

-0.14

OLF

-0.14

POSITIVE LOGITS

 pleasure

0.18

ffe

0.16

 partnerships

0.16

partner

0.16

iete

0.15

 affiliate

0.14

 Partner

0.14

ilan

0.14

propri

0.14

Activations Density 0.093%

phrases indicating the development or existence of programs and services

No Comments

No Known Activations