INDEX

Explanations

code structures, particularly functions and method definitions in programming languages

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

hur

-0.17

uard

-0.15

abh

-0.14

efon

-0.14

ergency

-0.14

ollider

-0.14

usher

-0.14

 tá»Ń

-0.14

.ACTION

-0.14

"""),↵

-0.14

POSITIVE LOGITS

0.23

}↵

0.20

0.17

 loose

0.16

};

0.16

 }↵

0.16

ences

0.15

}(

0.15

ibling

0.14

end

0.14

Activations Density 0.139%

code structures, particularly functions and method definitions in programming languages

No Comments

No Known Activations