INDEX

Explanations

terms related to drug use and addiction

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

ÅĦ

-0.16

eners

-0.15

aad

-0.15

omanip

-0.15

est

-0.15

ÏģÏī

-0.15

ãĥ£

-0.15

itore

-0.15

eniable

-0.14

consts

-0.14

POSITIVE LOGITS

store

0.25

stores

0.23

 abuse

0.23

 lord

0.23

 lords

0.21

 Lords

0.20

 dealer

0.20

OD

0.19

STORE

0.19

 possession

0.18

Activations Density 0.016%

terms related to drug use and addiction

No Comments

No Known Activations