INDEX

Explanations

references to numbers associated with forms or inputs

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 Demir

-0.16

 Tham

-0.15

eft

-0.15

is

-0.15

locking

-0.14

lap

-0.14

way

-0.14

 Reyes

-0.14

odes

-0.14

IGHL

-0.14

POSITIVE LOGITS

jÅ¡ÃŃ

0.16

.xtext

0.15

ynom

0.15

ailer

0.15

å¤§åħ¨

0.15

Wunused

0.14

Î²Îµ

0.14

neath

0.14

åįĪ

0.14

Activations Density 0.019%

references to numbers associated with forms or inputs

No Comments

No Known Activations