INDEX

Explanations

references to specific Lego themes and their historical context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

omor

-0.15

anÃ¡

-0.15

Thickness

-0.15

ija

-0.14

.Month

-0.14

 Ð±ÑĥÐº

-0.14

psy

-0.14

inox

-0.14

 depth

-0.14

aight

-0.13

POSITIVE LOGITS

bage

0.15

erness

0.15

_XML

0.15

 Fletcher

0.14

ulence

0.14

endoza

0.14

GORITH

0.14

ÃŃvel

0.14

.xml

0.13

acre

0.13

Activations Density 0.007%

references to specific Lego themes and their historical context

No Comments

No Known Activations