INDEX

Explanations

phrases or words indicating surprise or emphasis

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 10-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L10R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.10.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

ulg

-0.15

aviest

-0.15

sector

-0.14

arias

-0.14

ingles

-0.14

amble

-0.14

SingleOrDefault

-0.14

atin

-0.14

uala

-0.14

Dating

-0.14

POSITIVE LOGITS

.atom

0.14

lim

0.14

TO

0.13

rans

0.13

issan

0.13

_#{

0.13

 Energ

0.13

 Bram

0.13

 harness

0.13

ÏĩÎ±

0.12

Activations Density 0.005%

phrases or words indicating surprise or emphasis

No Comments

No Known Activations