INDEX

Explanations

locations and countries mentioned in the text

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

Negative Logits

inh

-0.15

ims

-0.14

ula

-0.14

mue

-0.14

Ø§Ø¦Ùĩ

-0.14

manuel

-0.14

legen

-0.13

aghetti

-0.13

oge

-0.13

.sol

-0.13

POSITIVE LOGITS

Macro

0.16

iyan

0.15

 tá»ĳ

0.14

Ð¾ÑĤ

0.14

 Howe

0.13

Ã©rÃ©

0.13

Ð¾Ð»Ñı

0.13

-US

0.13

ischer

0.13

rack

0.13

Activations Density 0.097%

locations and countries mentioned in the text

No Known Activations