INDEX

Explanations

terms related to advertising and ad technology

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

datatable

-0.15

bite

-0.15

uluk

-0.14

brook

-0.14

edImage

-0.14

_VIS

-0.13

eenth

-0.13

earch

-0.13

anton

-0.13

<fieldset

-0.13

POSITIVE LOGITS

words

0.27

Words

0.26

verts

0.24

verting

0.24

Sense

0.23

vertis

0.22

 sense

0.22

 serving

0.21

 Sense

0.20

WORDS

0.20

Activations Density 0.008%

terms related to advertising and ad technology

No Comments

No Known Activations