INDEX

Explanations

numerical values and their arrangements in various contexts

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 12-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L12R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.12.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

uls

-0.15

emi

-0.15

oken

-0.15

wayne

-0.15

oney

-0.14

arendra

-0.14

ptal

-0.14

alta

-0.14

ala

-0.14

POSITIVE LOGITS

uchos

0.16

ãĢģäºĮ

0.15

ãĥ¼ãĥŃ

0.15

 Rencontre

0.15

ãĢģä¸Ń

0.15

ë¶ĢíĦ°

0.15

ernet

0.15

nings

0.15

GeneratedValue

0.15

ustin

0.14

Activations Density 0.299%

numerical values and their arrangements in various contexts

No Comments

No Known Activations