INDEX

Explanations

negative terms describing foolishness or lack of intelligence

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

 Madness

-0.15

anford

-0.15

 badass

-0.14

chner

-0.14

mitted

-0.14

'er

-0.14

 madness

-0.14

iek

-0.14

onth

-0.13

anted

-0.13

POSITIVE LOGITS

ness

0.26

ities

0.24

ly

0.21

NESS

0.21

ity

0.19

ass

0.19

visor

0.18

ASS

0.17

-looking

0.17

/il

0.17

Activations Density 0.080%

negative terms describing foolishness or lack of intelligence

No Comments

No Known Activations