INDEX

Explanations

references to personal experiences and choices in a narrative context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

à¸µà¸¢à¸Ķ

-0.16

ault

-0.15

olg

-0.15

achuset

-0.14

ffi

-0.14

 Becker

-0.13

isel

-0.13

slu

-0.13

.createComponent

-0.13

ican

-0.13

POSITIVE LOGITS

çļĦæĺ¯

0.20

is

0.17

 belongs

0.17

aines

0.16

ìŀ¥ìĿĢ

0.16

isn

0.16

 actually

0.16

oÄŁ

0.16

nik

0.15

 tonight

0.15

Activations Density 0.156%

references to personal experiences and choices in a narrative context

No Comments

No Known Activations