INDEX

Explanations

HTML elements and formatting tags

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

ses

-0.16

ÐµÐ³

-0.15

TD

-0.15

ÑħÑĸÐ²

-0.14

 Rubio

-0.14

clave

-0.13

ÙıÙĪ

-0.13

ESS

-0.13

rat

-0.13

Ug

-0.13

POSITIVE LOGITS

ervas

0.18

fout

0.16

Rotor

0.15

andin

0.15

ToWorld

0.15

ukkit

0.14

Ka

0.14

VML

0.14

 åĮĹäº¬

0.14

weit

0.13

Activations Density 0.044%

HTML elements and formatting tags

No Comments

No Known Activations