INDEX

Explanations

specific health-related guidelines and recommendations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 6-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L6R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.6.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

SystemService

-0.18

pedia

-0.17

ÙĬØ©

-0.16

-0.14

moid

-0.14

Alice

-0.14

 Personality

-0.14

 Diseases

-0.14

 Alice

-0.14

eyer

-0.14

POSITIVE LOGITS

arna

0.16

swer

0.15

XL

0.14

 yard

0.14

HG

0.13

-yard

0.13

meni

0.13

è¨ĵ

0.12

-packages

0.12

 Yard

0.12

Activations Density 0.000%

specific health-related guidelines and recommendations

No Comments

No Known Activations