INDEX

Explanations

terms related to interactions and comments in a digital or social media context

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

æķ¦

-0.16

ucus

-0.15

ETYPE

-0.15

alian

-0.14

coni

-0.14

loff

-0.14

umar

-0.14

culus

-0.14

OSC

-0.14

-c

-0.14

POSITIVE LOGITS

_tF

0.15

NonQuery

0.15

Ã¤Ã¤n

0.14

ï¼²

0.14

obe

0.14

 fore

0.14

CCA

0.13

phan

0.13

ahr

0.13

.syn

0.13

Activations Density 0.048%

terms related to interactions and comments in a digital or social media context

No Comments

No Known Activations