INDEX

Explanations

variations of email and login references

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 19-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L19R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.19.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

 è°

-0.17

ance

-0.16

ili

-0.15

yn

-0.15

anzi

-0.15

uber

-0.15

.getProperties

-0.14

iversit

-0.14

 å·

-0.14

ange

-0.14

POSITIVE LOGITS

 protected

0.19

 Newsletter

0.15

 Cand

0.15

ãĥĲãĤ¹

0.14

Å¾e

0.14

 Sanct

0.14

 protective

0.14

 Sergio

0.14

rescia

0.14

Ron

0.14

Activations Density 0.004%

variations of email and login references

No Comments

No Known Activations