INDEX

Explanations

phrases related to documentation, review processes, and regulatory compliance

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 26-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L26R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.26.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

æĬĬ

-0.19

 deduct

-0.18

Generated

-0.17

conduct

-0.17

 GENERATED

-0.16

 tore

-0.16

 implant

-0.16

 eject

-0.16

 abort

-0.16

 withdrew

-0.16

POSITIVE LOGITS

 thrown

0.25

 given

0.25

 taken

0.22

put

0.20

 flown

0.19

 shown

0.19

 kept

0.18

 photographed

0.18

 nurs

0.18

 inventor

0.18

Activations Density 0.324%

phrases related to documentation, review processes, and regulatory compliance

No Comments

No Known Activations