INDEX

Explanations

instances of discussions and dialogues

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 24-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L24R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.24.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

ossip

-0.17

lem

-0.15

lemn

-0.15

-0.14

 communicate

-0.14

/DD

-0.14

 carp

-0.14

hlas

-0.14

unk

-0.14

ery

-0.14

POSITIVE LOGITS

 starter

0.27

 about

0.24

 starters

0.24

starter

0.24

 Starter

0.23

-about

0.20

_about

0.20

 About

0.19

ative

0.19

about

0.18

Activations Density 0.053%

instances of discussions and dialogues

No Comments

No Known Activations