INDEX

Explanations

references to null, undefined, or empty values in code-related contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

irc

-0.16

apro

-0.16

dip

-0.16

Dip

-0.15

astr

-0.14

esco

-0.14

ple

-0.14

oli

-0.14

mand

-0.14

mo

-0.14

POSITIVE LOGITS

 zero

0.22

 Empty

0.21

 Zero

0.20

-zero

0.20

.Empty

0.20

Empty

0.20

_empty

0.20

.empty

0.19

 EMPTY

0.18

-empty

0.18

Activations Density 0.149%

references to null, undefined, or empty values in code-related contexts

No Comments

No Known Activations