INDEX

Explanations

high numerical values associated with project components or configurations

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L1R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.1.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

idor

-0.16

.axis

-0.16

Trivia

-0.15

zew

-0.15

ESCO

-0.15

cmc

-0.14

curring

-0.14

svc

-0.14

iddy

-0.14

iske

-0.13

POSITIVE LOGITS

aji

0.15

¯

0.15

Î¼Î¯

0.14

hest

0.14

ÐĿÐĲ

0.14

urrent

0.14

 Dwight

0.14

old

0.13

uada

0.13

ombs

0.13

Activations Density 0.046%

high numerical values associated with project components or configurations

No Comments

No Known Activations