INDEX

Explanations

expressions of friendliness and supportive social interactions

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 21-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L21R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.21.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

ors

-0.20

CCA

-0.16

ÑĨÑİ

-0.15

åĸľ

-0.15

iled

-0.14

sf

-0.14

Ø³Ø§ÙĨ

-0.14

uiltin

-0.14

odzi

-0.14

POSITIVE LOGITS

lier

0.24

liness

0.21

 confines

0.20

liest

0.19

lies

0.18

ships

0.17

 disposed

0.17

 faces

0.17

ness

0.17

 enough

0.17

Activations Density 0.023%

expressions of friendliness and supportive social interactions

No Comments

No Known Activations