INDEX

Explanations

references to training and accreditation processes

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 25-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L25R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.25.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

ounce

-0.17

æķı

-0.14

otti

-0.14

zes

-0.14

ÑĥÑĪ

-0.14

æĭį

-0.14

 Dram

-0.14

Ð»Ð°Ð½

-0.14

unt

-0.13

ilio

-0.13

POSITIVE LOGITS

 scoped

0.18

 SHALL

0.18

 guidance

0.17

 scopes

0.17

 Dangerous

0.17

 Guidance

0.16

BS

0.16

 Generic

0.16

 generic

0.15

_upd

0.15

Activations Density 0.004%

references to training and accreditation processes

No Comments

No Known Activations