INDEX

Explanations

programming-related keywords and statements concerning policy validation and error handling

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 7-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L7R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.7.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

eros

-0.17

umbn

-0.16

à¥ĭà¤§

-0.16

Ð²Ð¸Ð³

-0.16

outil

-0.15

etest

-0.14

ãģ¡ãĤĥ

-0.14

 sort

-0.14

odb

-0.14

izards

-0.14

POSITIVE LOGITS

loff

0.16

hma

0.16

ptron

0.15

ackbar

0.15

ddy

0.14

Bea

0.14

 Advoc

0.14

ommen

0.14

 Regulation

0.14

wart

0.14

Activations Density 0.006%

programming-related keywords and statements concerning policy validation and error handling

No Comments

No Known Activations