INDEX

Explanations

articles, measurements, and descriptive phrases

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

New Auto-Interp

Top Features by Cosine Similarity

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

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L8R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.8.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

QP

-0.15

Ð·Ñĸ

-0.15

 prest

-0.14

pton

-0.14

uh

-0.14

 Herbert

-0.14

nost

-0.13

achen

-0.13

alytics

-0.13

POSITIVE LOGITS

 clearing

0.14

endforeach

0.14

 grading

0.14

ÑīÐ¸Ð½Ð°

0.14

 Innoc

0.13

apeut

0.13

ëŁ¬

0.13

Ð¾ÐºÑĥ

0.13

 Damian

0.13

cue

0.13

Activations Density 0.017%

articles, measurements, and descriptive phrases

No Comments

No Known Activations