INDEX

Explanations

references to news articles, particularly those detailing events or topics from Colorado

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

ispecies

-0.15

oÄį

-0.14

lÃŃÄį

-0.14

coles

-0.14

ÑĤÑı

-0.14

_mt

-0.14

lassen

-0.14

á»Ļi

-0.14

 creep

-0.13

POSITIVE LOGITS

utra

0.17

ãĤ¤ãĥī

0.16

ph

0.16

 caus

0.15

ä¿

0.14

 shel

0.14

 Conduct

0.14

hi

0.14

 corners

0.14

Activations Density 0.014%

references to news articles, particularly those detailing events or topics from Colorado

No Comments

No Known Activations