INDEX

Explanations

proper nouns, particularly names and titles

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 31-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L31R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.31.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

 (Â«

-0.16

 ï¼į

-0.14

Â´s

-0.14

 Cush

-0.13

—

-0.13

 denn

-0.13

amen

-0.13

Â«

-0.13

atus

-0.13

ISBN

-0.13

POSITIVE LOGITS

 kazan

0.23

zam

0.19

kaz

0.15

kad

0.15

Hz

0.14

 curiosity

0.14

Ì§

0.14

Kad

0.14

_visitor

0.14

'il

0.13

Activations Density 0.005%

proper nouns, particularly names and titles

No Comments

No Known Activations