INDEX

Explanations

timestamps and references related to digital media or online communications

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

vg

-0.19

fg

-0.15

ACHI

-0.15

 carrying

-0.15

 spreads

-0.14

iate

-0.14

vm

-0.14

atoi

-0.14

ÏĩÎ¿

-0.14

Ð¾Ð²Ð¾

-0.13

POSITIVE LOGITS

reen

0.16

pNet

0.16

.GraphicsUnit

0.15

ÐºÐ¾ÑĢÐ¸ÑģÑĤ

0.15

nels

0.15

.Cursors

0.14

LinkId

0.14

leDb

0.14

BarItem

0.14

underscore

0.14

Activations Density 0.008%

timestamps and references related to digital media or online communications

No Comments

No Known Activations