INDEX

Explanations

proper nouns, specifically names of people and organizations

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

New Auto-Interp

Top Features by Cosine Similarity

Comparing With LLAMA3.1-8B @ 16-llamascope-res-32k

Configuration

fnlp/Llama3_1-8B-Base-LXR-8x/Llama3_1-8B-Base-L16R-8x

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

cerebras/SlimPajama-627B

Features

32,768

Data Type

bfloat16

Hook Name

blocks.16.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

atura

-0.17

 conf

-0.15

atri

-0.15

uco

-0.15

Ort

-0.14

entina

-0.14

robat

-0.14

UGHT

-0.14

ittle

-0.14

orp

-0.14

POSITIVE LOGITS

Reporting

0.21

 Reporting

0.19

 reporting

0.18

ToDevice

0.14

uled

0.14

æĬ¥éģĵ

0.14

 reporters

0.13

 reporter

0.13

-report

0.13

Activations Density 0.033%

proper nouns, specifically names of people and organizations

No Comments

No Known Activations