INDEX

Explanations

references to resource utilization or searches for individuals

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

igh

-0.17

\Db

-0.17

cura

-0.15

antha

-0.15

çĲ´

-0.15

 Maid

-0.15

iro

-0.14

ÄĻk

-0.14

askell

-0.14

çµ

-0.14

POSITIVE LOGITS

 date

0.18

 proper

0.17

 above

0.17

 profile

0.16

 edit

0.16

 Date

0.15

 prof

0.15

.AutoComplete

0.15

above

0.15

elling

0.15

Activations Density 0.002%

references to resource utilization or searches for individuals

No Comments

No Known Activations