INDEX

Explanations

phrases related to reviewing and revisiting past content or experiences

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

irit

-0.17

ker

-0.16

iel

-0.15

çĽ¸

-0.15

berger

-0.15

bert

-0.14

olin

-0.14

erties

-0.14

uit

-0.14

ped

-0.14

POSITIVE LOGITS

 again

0.18

again

0.15

isc

0.14

asio

0.14

 Ð¿Ð¾Ð²ÑĤÐ¾ÑĢ

0.14

isiyle

0.14

ç§Ł

0.14

.Automation

0.14

_different

0.14

PUR

0.14

Activations Density 0.153%

phrases related to reviewing and revisiting past content or experiences

No Comments

No Known Activations