INDEX

Explanations

references to specific items or topics being discussed

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 9-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L9R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.9.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

 WithEvents

-0.15

 Restr

-0.15

 crossover

-0.15

 Ð¼Ð¾Ð»

-0.14

less

-0.14

ence

-0.14

phans

-0.14

Injected

-0.14

 tension

-0.14

ëĿ½

-0.14

POSITIVE LOGITS

acular

0.17

enaire

0.15

mdp

0.15

aban

0.14

Ð¸Ð½Ð¾Ð¹

0.14

otton

0.14

á»ĵi

0.13

ubbo

0.13

ÑĢÐ¸Ð²

0.13

kn

0.13

Activations Density 0.147%

references to specific items or topics being discussed

No Comments

No Known Activations