INDEX

Explanations

references to educational institutions and related terminology

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

igham

-0.16

assen

-0.15

edd

-0.14

 Hydra

-0.14

uire

-0.14

fy

-0.14

-Clause

-0.14

oppable

-0.14

awy

-0.14

POSITIVE LOGITS

yard

0.22

mates

0.20

children

0.20

astic

0.20

æ³ķäºº

0.19

 dropout

0.17

/work

0.16

house

0.16

 Dropout

0.16

/job

0.15

Activations Density 0.036%

references to educational institutions and related terminology

No Comments

No Known Activations