INDEX

Explanations

phrases indicating similarity or comparison between entities

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

iman

-0.16

æŃ¯

-0.15

angen

-0.14

é³

-0.14

Dal

-0.14

.ASCII

-0.13

dae

-0.13

ems

-0.13

rej

-0.13

essen

-0.13

POSITIVE LOGITS

 except

0.16

äº

0.15

 regular

0.15

ä¸Ģæł·

0.14

ancel

0.14

_PT

0.14

isiyle

0.14

except

0.14

asin

0.14

pth

0.14

Activations Density 0.109%

phrases indicating similarity or comparison between entities

No Comments

No Known Activations