INDEX

Explanations

references to art events and exhibitions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 29-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L29R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.29.hook_resid_post

Hook Layer

Architecture

jumprelu

Context Size

1,024

Dataset

cerebras/SlimPajama-627B

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

achuset

-0.16

 Kaynak

-0.16

uckle

-0.15

gewater

-0.14

 Vance

-0.14

ì¹´ì§Ģëħ¸

-0.14

ateau

-0.14

 Psychic

-0.14

_THRESH

-0.13

wr

-0.13

POSITIVE LOGITS

 Reload

0.20

&W

0.20

&P

0.20

&M

0.19

+:

0.19

&B

0.19

Lab

0.17

&T

0.17

Days

0.17

XS

0.17

Activations Density 0.408%

references to art events and exhibitions

No Comments

No Known Activations