INDEX

Explanations

user agent strings from web browsers

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 14-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L14R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.14.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

erdem

-0.17

ivol

-0.15

usz

-0.15

eday

-0.14

idis

-0.14

isbury

-0.14

zek

-0.14

erif

-0.14

.ecore

-0.14

ivant

-0.14

POSITIVE LOGITS

åĿĬ

0.15

tin

0.14

/parser

0.14

åª

0.13

é»İ

0.13

ParameterValue

0.13

Triangles

0.13

 Outline

0.13

 Lorenzo

0.12

IENT

0.12

Activations Density 0.014%

user agent strings from web browsers

No Comments

No Known Activations