INDEX

Explanations

numeric values and their relevance to particular events or topics

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 5-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L5R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.5.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

erb

-0.15

Ã¼re

-0.15

iltr

-0.14

riz

-0.14

PG

-0.14

.ov

-0.14

inch

-0.14

zym

-0.14

 Suspension

-0.13

Ã¼le

-0.13

POSITIVE LOGITS

uci

0.16

 ÙħÛĮÙĦØ§Ø¯ÛĮ

0.15

 Manor

0.15

PropTypes

0.14

Ã¨n

0.14

acula

0.14

LookAndFeel

0.14

 trend

0.14

 !***

0.14

ÑĥÐ´Ð¸

0.14

Activations Density 0.019%

numeric values and their relevance to particular events or topics

No Comments

No Known Activations