INDEX

Explanations

terms and concepts related to mathematical categories and structures

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 15-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L15R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.15.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

hangi

-0.16

agues

-0.15

LI

-0.14

UNET

-0.14

ÑįÑĤ

-0.14

 Parcel

-0.14

itel

-0.14

htub

-0.14

umbnail

-0.14

ITERAL

-0.14

POSITIVE LOGITS

Concurrency

0.15

iá»ĥn

0.15

/^\

0.14

jo

0.14

unma

0.14

 spaces

0.14

auty

0.14

0.13

 Junk

0.13

æĹĹ

0.13

Activations Density 0.174%

terms and concepts related to mathematical categories and structures

No Comments

No Known Activations