INDEX

Explanations

references to vintage or antique items

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

elter

-0.15

idon

-0.14

herits

-0.14

 histories

-0.14

Artist

-0.14

room

-0.14

 chocol

-0.14

-0.13

 ÙĪÛĮÚ©ÛĮ

-0.13

POSITIVE LOGITS

/class

0.24

-inspired

0.22

-style

0.22

grade

0.22

/mod

0.20

-looking

0.20

actively

0.18

/new

0.18

gressive

0.16

/rec

0.16

Activations Density 0.021%

references to vintage or antique items

No Comments

No Known Activations