INDEX

Explanations

terms related to health monitoring features

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

äº

-0.15

ucch

-0.15

[section

-0.15

ManagedObject

-0.14

 Lodge

-0.14

alli

-0.14

ERGY

-0.14

sek

-0.14

ODO

-0.14

POSITIVE LOGITS

DBG

0.14

Ð¾Ð²Ð¾ÑĢ

0.14

itos

0.14

utes

0.14

icap

0.14

APP

0.14

cea

0.13

aster

0.13

uhn

0.13

bate

0.13

Activations Density 0.045%

terms related to health monitoring features

No Comments

No Known Activations