INDEX

Explanations

comparisons and specific details related to behavior and education

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

Ø§Ø®

-0.16

Å¡Å¥

-0.14

zhou

-0.14

edith

-0.14

AINER

-0.14

atz

-0.13

 faint

-0.13

hou

-0.13

ideos

-0.13

antino

-0.13

POSITIVE LOGITS

 individual

0.32

 specific

0.29

åħ·ä½ĵ

0.27

individual

0.25

 ÐºÐ¾Ð½ÐºÑĢÐµÑĤ

0.24

 Specific

0.23

-specific

0.23

 Individual

0.23

 particular

0.23

specific

0.23

Activations Density 0.287%

comparisons and specific details related to behavior and education

No Comments

No Known Activations