INDEX

Explanations

terms related to fitness and exercise

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

ings

-0.18

StringEncoding

-0.16

ÙħØ§

-0.16

ervers

-0.16

Ã³c

-0.15

ingen

-0.15

TestingModule

-0.15

erville

-0.15

asso

-0.15

_TestCase

-0.14

POSITIVE LOGITS

readcr

0.18

-conscious

0.18

-minded

0.17

 center

0.17

Ø¯

0.16

usra

0.16

gram

0.15

Hub

0.15

 dove

0.15

 minded

0.15

Activations Density 0.023%

terms related to fitness and exercise

No Comments

No Known Activations