INDEX

Explanations

words and phrases that convey friendliness and positive 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 @ 8-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L8R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.8.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

Ìĥ

-0.16

ãĤ¤ãĤ¯

-0.15

cpp

-0.15

ngo

-0.14

 greed

-0.14

nam

-0.14

oen

-0.14

iao

-0.14

.Throw

-0.13

ngth

-0.13

POSITIVE LOGITS

inkel

0.19

assen

0.18

yyyy

0.16

 enough

0.15

iswa

0.15

liness

0.15

lier

0.15

ãĥ¶

0.15

ness

0.14

 WithEvents

0.14

Activations Density 0.023%

words and phrases that convey friendliness and positive social interactions

No Comments

No Known Activations