INDEX

Explanations

words indicating the presence or condition of various subjects and actions related to status and existence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.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

lyph

-0.18

ï»¿↵↵

-0.16

ÎŃÎ»

-0.15

ï»¿using

-0.14

ofire

-0.14

.createFrom

-0.14

ekim

-0.14

oader

-0.14

'gc

-0.14

istrovstvÃŃ

-0.14

POSITIVE LOGITS

 maid

0.21

 regist

0.20

 build

0.19

 feed

0.19

 setup

0.19

 hold

0.18

 bind

0.18

 send

0.18

 drown

0.17

 finish

0.17

Activations Density 0.463%

words indicating the presence or condition of various subjects and actions related to status and existence

No Comments

No Known Activations