INDEX

Explanations

expressions of expectation or surprise related to personal achievements or experiences

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 18-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L18R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.18.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

chten

-0.16

buie

-0.15

elligent

-0.15

_pas

-0.14

ulary

-0.14

.scalablytyped

-0.14

çĦ¶

-0.14

Ø±Ø§ÙĨÙĩ

-0.14

è¡Ĩ

-0.14

cmc

-0.14

POSITIVE LOGITS

 Little

0.24

æĸĻ

0.21

Little

0.21

 Expect

0.20

 little

0.19

 expectations

0.19

wil

0.19

 expected

0.18

expect

0.18

Expect

0.17

Activations Density 0.110%

expressions of expectation or surprise related to personal achievements or experiences

No Comments

No Known Activations