INDEX

Explanations

phrases that indicate sources of additional information and resources

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

-0.15

ado

-0.14

.sm

-0.14

elman

-0.14

illin

-0.14

uc

-0.14

icer

-0.13

ikh

-0.13

ows

-0.13

POSITIVE LOGITS

rimp

0.15

makt

0.15

Sou

0.14

ĤŃ

0.14

stva

0.14

.mvc

0.14

 fitte

0.13

autÃ©

0.13

unta

0.13

à¹ģà¸ļà¸ļ

0.13

Activations Density 0.061%

phrases that indicate sources of additional information and resources

No Comments

No Known Activations