INDEX

Explanations

terms related to Gothic horror

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 4-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L4R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.4.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

 Evil

-0.16

yr

-0.15

 vice

-0.15

Dra

-0.15

 evil

-0.15

APPLE

-0.15

evil

-0.14

vice

-0.14

 Vice

-0.14

 apple

-0.14

POSITIVE LOGITS

erner

0.16

/Typography

0.16

enth

0.16

 Seas

0.15

ROP

0.15

bolt

0.14

WithData

0.14

aseline

0.14

_literals

0.14

servers

0.14

Activations Density 0.009%

terms related to Gothic horror

No Comments

No Known Activations