INDEX

Explanations

the word "inside" and its various contexts in connection with buildings or structures

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

eros

-0.17

ataires

-0.17

mgr

-0.16

erus

-0.16

ws

-0.15

ched

-0.15

eness

-0.15

dll

-0.15

ëĿ½

-0.15

thing

-0.15

POSITIVE LOGITS

/out

0.40

-out

0.27

 ÙĪØ®

0.23

-Out

0.22

/on

0.22

halb

0.21

OUT

0.21

/up

0.20

wards

0.20

Out

0.19

Activations Density 0.033%

the word "inside" and its various contexts in connection with buildings or structures

No Comments

No Known Activations