INDEX

Explanations

instances of the word "form" and its variations in different contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ements

-0.17

patch

-0.15

.raise

-0.14

educt

-0.14

Gra

-0.14

orra

-0.14

orca

-0.14

at

-0.14

istro

-0.14

POSITIVE LOGITS

idable

0.24

ulary

0.22

ulas

0.22

ulating

0.21

osa

0.21

ateur

0.20

-urlencoded

0.19

ality

0.19

ulators

0.19

ULA

0.18

Activations Density 0.022%

instances of the word "form" and its variations in different contexts

No Comments

No Known Activations