INDEX

Explanations

references to specific characteristics or features of items and their impact or relationship to various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 11-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L11R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.11.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

hook

-0.15

rese

-0.15

 à¸¿

-0.15

volution

-0.14

 Sanayi

-0.14

 ê°Ī

-0.13

rk

-0.13

ificial

-0.13

 pole

-0.13

à¹īà¸Ńà¸¢

-0.13

POSITIVE LOGITS

eland

0.18

iol

0.17

osl

0.15

orpor

0.15

onis

0.15

HeaderValue

0.15

PEND

0.15

 intens

0.14

sam

0.14

antan

0.14

Activations Density 0.056%

references to specific characteristics or features of items and their impact or relationship to various contexts

No Comments

No Known Activations