INDEX

Explanations

specific names or titles, particularly those that seem to represent people or characters in various contexts

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

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

_DIST

-0.15

egot

-0.14

ximity

-0.14

wood

-0.13

ioc

-0.13

aney

-0.13

dal

-0.13

virt

-0.13

egov

-0.13

æ¸¡

-0.13

POSITIVE LOGITS

ertest

0.17

isia

0.16

=pk

0.15

*)((

0.15

nown

0.15

orks

0.14

íİ¸

0.14

orgot

0.14

ool

0.13

ylon

0.13

Activations Density 0.503%

specific names or titles, particularly those that seem to represent people or characters in various contexts

No Comments

No Known Activations