INDEX

Explanations

phrases and terms related to health and wellness treatments

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

 eval

-0.15

orer

-0.15

RTL

-0.15

 abstract

-0.15

 cust

-0.15

 Cust

-0.14

 Agent

-0.14

çĶĺ

-0.14

ocio

-0.14

 agent

-0.14

POSITIVE LOGITS

 Harley

0.27

 clinic

0.24

 treatments

0.24

 Clinic

0.23

 clinics

0.23

clinic

0.20

 Consultant

0.20

GP

0.20

GP

0.20

 Treatment

0.18

Activations Density 0.053%

phrases and terms related to health and wellness treatments

No Comments

No Known Activations