INDEX

Explanations

phrases that describe a collection or grouping of items or standards

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

edn

-0.20

enthal

-0.19

ishly

-0.16

usu

-0.14

ĮĴ

-0.14

ãĤ¤ãĥĦ

-0.14

uner

-0.14

enburg

-0.14

atern

-0.14

aden

-0.14

POSITIVE LOGITS

tle

0.23

 aside

0.22

sockopt

0.22

osa

0.21

tlement

0.21

jmp

0.21

 forth

0.20

metatable

0.19

-up

0.19

aside

0.19

Activations Density 0.030%

phrases that describe a collection or grouping of items or standards

No Comments

No Known Activations