INDEX

Explanations

references to events, actions, and interactions involving people and places

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

 Mund

-0.15

æ»ħ

-0.15

?action

-0.15

 çĿ

-0.15

argent

-0.15

bury

-0.15

azer

-0.14

gul

-0.14

amento

-0.13

POSITIVE LOGITS

adol

0.14

alars

0.14

 Brick

0.14

ocha

0.14

itious

0.14

akat

0.14

_Utils

0.14

atan

0.14

Leaks

0.13

atak

0.13

Activations Density 0.011%

references to events, actions, and interactions involving people and places

No Comments

No Known Activations