INDEX

Explanations

references to the interaction between alcohol and medications or drugs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

 orth

-0.15

 Candy

-0.14

olini

-0.14

 Lunch

-0.13

din

-0.13

agh

-0.13

ickle

-0.13

 hungry

-0.13

 electrical

-0.13

 Breakfast

-0.13

POSITIVE LOGITS

 alcohol

0.60

 Alcohol

0.52

éħĴ

0.51

 alcoholic

0.50

 drink

0.48

 drinking

0.46

 booze

0.46

 Ð°Ð»ÐºÐ¾Ð³

0.45

 drunk

0.44

 rÆ°á»£u

0.44

Activations Density 0.306%

references to the interaction between alcohol and medications or drugs

No Comments

No Known Activations