INDEX

Explanations

ordinal indicators, particularly the suffix "th" and other ordinal markers

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

hip

-0.16

het

-0.15

 duty

-0.14

ierz

-0.14

__

-0.14

opal

-0.14

Yuk

-0.14

bar

-0.14

onder

-0.14

eller

-0.13

POSITIVE LOGITS

ensen

0.16

 ÙħÛĮÙĦØ§Ø¯ÛĮ

0.15

Ä±s

0.15

/*č↵

0.14

lander

0.14

Ã¼zel

0.14

 Bryant

0.14

 Cyril

0.14

/TT

0.14

å¯¸

0.14

Activations Density 0.014%

ordinal indicators, particularly the suffix "th" and other ordinal markers

No Comments

No Known Activations