INDEX

Explanations

references to the concept of "state" and its variations within the text

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

_states

-0.21

hey

-0.21

thon

-0.20

andes

-0.18

them

-0.18

StateChanged

-0.17

 stating

-0.17

Statement

-0.17

ulk

-0.17

teenth

-0.17

POSITIVE LOGITS

craft

0.32

hood

0.28

-of

0.23

 Unidos

0.22

ful

0.22

fully

0.22

house

0.19

coach

0.19

MENT

0.19

MENTS

0.19

Activations Density 0.090%

references to the concept of "state" and its variations within the text

No Comments

No Known Activations