INDEX

Explanations

various punctuation marks and formatting features related to dates and times

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

åĵ

-0.15

hed

-0.15

olic

-0.15

abis

-0.15

ria

-0.15

ilda

-0.15

VL

-0.14

Hed

-0.14

 Hour

-0.14

BAB

-0.14

POSITIVE LOGITS

 Enemy

0.17

(MPI

0.15

grade

0.15

oldown

0.15

 grade

0.15

ewire

0.15

Enemy

0.14

 jadx

0.14

ToObject

0.14

éº

0.14

Activations Density 0.004%

various punctuation marks and formatting features related to dates and times

No Comments

No Known Activations