INDEX

Explanations

terms related to disabilities and recovery processes

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

itas

-0.17

.transition

-0.15

inos

-0.15

consts

-0.14

lul

-0.14

steen

-0.14

ahoo

-0.14

ktop

-0.14

hea

-0.14

sut

-0.14

POSITIVE LOGITS

 paralysis

0.30

 motor

0.29

 Motor

0.27

 paralyzed

0.27

motor

0.27

 MOTOR

0.25

Motor

0.24

_motor

0.22

 speech

0.21

 motors

0.21

Activations Density 0.034%

terms related to disabilities and recovery processes

No Comments

No Known Activations