INDEX

Explanations

the name "Andrew" and its variations in the text

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

erap

-0.16

opath

-0.15

íĴį

-0.15

ibur

-0.15

erosis

-0.15

 sscanf

-0.15

ITO

-0.15

ern

-0.14

STANCE

-0.14

.scalablytyped

-0.14

POSITIVE LOGITS

 Cuomo

0.18

 Carnegie

0.17

 Mellon

0.17

ollo

0.15

ski

0.15

ship

0.15

sup

0.14

ed

0.14

REFERRED

0.14

Activations Density 0.008%

the name "Andrew" and its variations in the text

No Comments

No Known Activations