INDEX

Explanations

code snippets related to function definitions and control structures

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

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

vit

-0.15

esser

-0.15

rait

-0.15

essen

-0.14

ster

-0.14

nbsp

-0.14

ulo

-0.14

lar

-0.13

ara

-0.13

POSITIVE LOGITS

affiliate

0.15

toPromise

0.14

 puck

0.14

ypo

0.14

ÙĨØ¯ÙĬ

0.14

entiful

0.13

PasswordField

0.13

inction

0.13

uffix

0.13

Orc

0.13

Activations Density 0.077%

code snippets related to function definitions and control structures

No Comments

No Known Activations