INDEX

Explanations

references to "education" and related academic terms

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

ao

-0.17

acha

-0.16

logy

-0.14

ocular

-0.14

oney

-0.14

 Yates

-0.14

sten

-0.13

lege

-0.13

umatic

-0.13

iding

-0.13

POSITIVE LOGITS

ardown

0.16

imple

0.16

ally

0.16

ritic

0.15

Č

0.15

.scalablytyped

0.15

IALOG

0.15

ationally

0.15

mund

0.15

ALLY

0.14

Activations Density 0.038%

references to "education" and related academic terms

No Comments

No Known Activations