INDEX

Explanations

word forms related to the concept of "end" or "ending."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 2-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L2R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.2.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

oder

-0.16

oga

-0.15

aa

-0.14

Ã¨ge

-0.14

nce

-0.14

pn

-0.14

oint

-0.13

 Ders

-0.13

ando

-0.13

odo

-0.13

POSITIVE LOGITS

OutOfBounds

0.20

hawk

0.16

egl

0.15

ÐºÐ°Ð²

0.15

IMS

0.14

ergy

0.14

 thumbs

0.14

 crossorigin

0.14

ellite

0.14

forge

0.14

Activations Density 0.009%

word forms related to the concept of "end" or "ending."

No Comments

No Known Activations