INDEX

Explanations

instances of the word "was."

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 20-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L20R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.20.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

esser

-0.18

gon

-0.15

RESSED

-0.15

lech

-0.15

ATAR

-0.14

æŃ©

-0.14

ovali

-0.14

iyel

-0.14

cri

-0.14

ondon

-0.14

POSITIVE LOGITS

abi

0.23

ps

0.17

ier

0.16

Ç

0.16

illa

0.16

ows

0.16

ist

0.15

imate

0.15

ient

0.14

ly

0.14

Activations Density 0.135%

instances of the word "was."

No Comments

No Known Activations