INDEX

Explanations

terms related to health-related content and downloadable resources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

angs

-0.15

onga

-0.13

jad

-0.13

Gig

-0.13

.cpu

-0.13

 ç²¾

-0.13

olas

-0.13

 Gard

-0.13

ÐºÑĢÐµÑĤ

-0.13

ped

-0.12

POSITIVE LOGITS

 Cotton

0.15

ĥģ

0.15

rieg

0.15

usan

0.14

ennen

0.14

 flatt

0.14

iddi

0.14

ä

0.14

ACK

0.13

illance

0.13

Activations Density 0.015%

terms related to health-related content and downloadable resources

No Comments

No Known Activations