INDEX

Explanations

punctuation marks and formatting elements commonly found in written text

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

aran

-0.16

gro

-0.16

ien

-0.15

arc

-0.15

mus

-0.15

otropic

-0.14

rod

-0.14

end

-0.14

mes

-0.14

yan

-0.14

POSITIVE LOGITS

 CURLOPT

0.17

Ð¶Ð´

0.15

icine

0.15

ŀæĢ§

0.15

rire

0.14

ASSERT

0.14

MethodName

0.14

éĶ

0.14

zahl

0.14

ecycle

0.13

Activations Density 0.004%

punctuation marks and formatting elements commonly found in written text

No Comments

No Known Activations