INDEX

Explanations

variations of the word "prom."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

irst

-0.16

iculty

-0.16

zig

-0.15

.Decode

-0.15

vre

-0.15

irut

-0.15

riangle

-0.14

arges

-0.14

letion

-0.14

raud

-0.14

POSITIVE LOGITS

etheus

0.25

 prom

0.22

 Prom

0.20

enade

0.19

(prom

0.18

 PROM

0.17

/prom

0.17

 Closet

0.16

Prom

0.16

ises

0.15

Activations Density 0.017%

variations of the word "prom."

No Comments

No Known Activations