INDEX

Explanations

references to different nodes within a system or network

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 13-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L13R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.13.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

ner

-0.19

mente

-0.19

ness

-0.17

-minded

-0.17

bourg

-0.17

ly

-0.17

bred

-0.16

nga

-0.15

teenth

-0.15

spe

-0.15

POSITIVE LOGITS

fault

0.19

uality

0.17

elli

0.16

ito

0.16

/system

0.16

ules

0.16

éĹ´

0.15

æķ°

0.15

hood

0.14

upon

0.14

Activations Density 0.073%

references to different nodes within a system or network

No Comments

No Known Activations