INDEX

Explanations

phrases indicating conditions and expectations in software or policy validation contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

_extended

-0.16

Ìī

-0.14

Î¹ÎºÎ±

-0.14

od

-0.13

 Rape

-0.13

ADB

-0.13

 enclosure

-0.13

 sinc

-0.13

lessly

-0.13

fort

-0.13

POSITIVE LOGITS

Ð²Ð°Ð¹

0.14

igham

0.14

emies

0.13

çŃĴ

0.13

#index

0.13

HasForeignKey

0.13

plusplus

0.13

isman

0.13

_NC

0.13

rlen

0.13

Activations Density 0.019%

phrases indicating conditions and expectations in software or policy validation contexts

No Comments

No Known Activations