INDEX

Explanations

parentheses in the text

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 23-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L23R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.23.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

-0.16

inger

-0.14

 Deputy

-0.14

 basket

-0.14

or

-0.14

 supposed

-0.13

umd

-0.13

.class

-0.13

 Cave

-0.13

POSITIVE LOGITS

AP

0.19

SEND

0.16

Reuters

0.15

reesome

0.15

ElementsBy

0.15

BUS

0.14

Ä°K

0.14

loat

0.14

aliyet

0.14

press

0.14

Activations Density 0.007%

parentheses in the text

No Comments

No Known Activations