INDEX

Explanations

quantitative references, particularly numbers and numerical 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

iaux

-0.20

_dropout

-0.15

etty

-0.14

ï»¿↵↵

-0.14

Ø¬Ø¹

-0.14

Ð¿Ð¾ÑĢ

-0.14

kowski

-0.14

UPLE

-0.14

raith

-0.13

Ð¾ÑĢÐ¾Ð½

-0.13

POSITIVE LOGITS

ugin

0.16

 mutant

0.14

Vic

0.14

chy

0.14

 sized

0.14

roke

0.13

iot

0.13

 Reeves

0.13

ÑĨÑĭ

0.13

iana

0.13

Activations Density 0.075%

quantitative references, particularly numbers and numerical phrases

No Comments

No Known Activations