INDEX

Explanations

personal pronouns and expressions of intention

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 27-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L27R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.27.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

 âĸ²

-0.15

 rubbish

-0.15

ije

-0.15

lege

-0.14

ouce

-0.14

leans

-0.14

iji

-0.13

ÙĤÙĩ

-0.13

onta

-0.13

bab

-0.13

POSITIVE LOGITS

 recently

0.21

 faced

0.21

 facing

0.20

faces

0.18

face

0.18

 have

0.18

 Recently

0.17

éĿ¢

0.17

 faces

0.17

 trying

0.17

Activations Density 0.064%

personal pronouns and expressions of intention

No Comments

No Known Activations