INDEX

Explanations

names and terms associated with specific individuals or characters

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 22-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L22R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.22.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

isman

-0.15

lag

-0.15

owski

-0.14

ãĥ¼ãĥĩ

-0.13

Mahon

-0.13

 chance

-0.13

endency

-0.13

Ð°ÑĢÐ°Ð½

-0.13

builder

-0.13

otti

-0.13

POSITIVE LOGITS

iflower

0.23

chwitz

0.19

inand

0.17

auga

0.16

itz

0.16

0.15

erer

0.15

 chuyá»ĩn

0.14

trag

0.14

quier

0.14

Activations Density 0.045%

names and terms associated with specific individuals or characters

No Comments

No Known Activations