INDEX

Explanations

conversational elements and dialogue

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 30-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L30R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.30.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

 Indeed

-0.20

 indeed

-0.19

 given

-0.19

 Simply

-0.19

 whilst

-0.18

 upon

-0.17

Indeed

-0.17

 simply

-0.17

 Thus

-0.17

Simply

-0.17

POSITIVE LOGITS

 People

0.23

 people

0.22

They

0.22

 everybody

0.22

People

0.22

they

0.21

 They

0.21

people

0.21

 they

0.21

when

0.19

Activations Density 0.849%

conversational elements and dialogue

No Comments

No Known Activations

conversational elements and dialogue

No Comments

No Known Activations